VoIP Myth Busters

VoIP Myth Busters

Using internet technology to deliver voice service has finally become ready for prime time. Learn the truths behind the myths surrounding this technology--and why more institutions are signing on.

Five years ago, a technology professional couldn't turn around in a crowded room without bumping into a vendor selling a hot new technology called Voice over Internet Protocol (VoIP). Each year brought the same promises of how VoIP would revolutionize the delivery of phone service, replacing expensive and cumbersome traditional phone service delivered by the "Baby Bells" with a cheap alternative.

 

Better sound quality was promised, but the reality was static and choppy connections. So most organizations wisely stayed away from the new technology.

Things change, but the college or university now considering adapting VoIP technology still confronts a series of popular myths. Here, University Business debunks the top 10 myths and uncovers the realities of a technology that has finally begun to be deployed on campuses large and small.

 

Reality: Implementation likely requires outside help, and IT staffers had best be spending their time learning how their jobs will be changing after VoIP comes to campus.

Moving from a separate voice-based phone system to one in which voice and data are shared over the data network does eliminate some complexity from the systems. However, the new hybrid network will require some new skills, and therefore a degree of retraining of IT staff may be necessary.

With VoIP, "you get the synergy of being able to use your existing Ethernet network and fiber optic networks to provide telephony," says John Bryan, vice president for Information Technology and Services at Clayton State University in Morrow, Ga. When tech leaders there adopted IP telephony, they didn't have to do much new hiring and training. In fact, Bryan notes, "We've been able to eliminate one whole type of technology, and that's a tremendous savings in terms of cost and support."

Though they understand the technology, IT staff typically don't have voice expertise, explains Todd Grafton, an engineer with CDW Government, a technology products and services provider. A person trained in data networks may not think to ask certain questions, such as how direct-dial numbers might affect the arrangement. But most vendors will provide training courses to their customers as part of the implementation process.

Chip Towle, senior IT director at Boston's Wheelock College, says the amount of retraining needed for his campus' switchover was less of a problem than ensuring that the campus' Nortel data network was properly configured to support voice as an application. In March 2003, he deployed VoIP phones, and today 95 percent of Wheelock's faculty and staff use VoIP. Training presented no problem, and Towle reports that he has no traditional phone technology staff on campus.

Reality: VoIP can be less expensive than traditional phone service, but free it is not.

"Every now and then, somebody is under the impression that if they deploy a VoIP solution for their office, they will end up with free phone service," says Grafton. That is not true, he adds, but the "base price of a VoIP platform is vendor-determined; there are more expensive and cheaper ones." Also, a lot of institutions of higher education are finding it makes sense to switch to VoIP as part of a larger overhaul of their data networks, which blurs the distinctions about what costs are for what purpose.

But that's not to say that there are not some significant savings from VoIP once it has been installed.

A return on investment for Clayton State's VoIP system will take five years, reports Bryan, and better yet, ongoing operating costs are much lower. He says that after five years, the VoIP system will cost half of what it would have cost to keep operating a traditional phone service, known as plain old telephone service (POTS).

Russ Beard, director of Information and Communication Services at Big Bend Community College in Moses Lake, Wash., estimates that his campus' VoIP system costs are "almost identical" to the institution's former phone system. Yet, when he adds in all of the services and features (such as caller ID and voicemail) that are standard in VoIP, the balance tips in favor of the newer technology.

Towle finds VoIP to be much cheaper, in part because when he upgrades one part of his network, he is upgrading his data and voice application capacity at the same time. "Voice is [simply another] application running on a data network today," he says.

The changing economics of the telephony industry are also working to the advantage of VoIP systems. With the increased competition in the telecommunications industry since it was deregulated, users can get competitive bids and lower prices for services.

"Long distance used to be a profit center in higher education," says Bryan. "When there was no competition and the phone companies were regulated, the economic model made sense to have centralized phone companies that had very expensive switches to provide expensive long-distance and local services. It was long-distance services that many colleges and universities resold to their students and [departments] and made a profit on, so they could afford to buy a large switch and get the economies of scale to run it themselves."

But since the mid-1990s, long-distance costs have declined dramatically. Calls that once cost 25 or 50 cents a minute dropped to as little as 1 cent a minute if bought in bulk. In this new paradigm, Clayton State no longer charges back to departments the costs of long distance. Billing and account maintenance costs simply "don't make sense economically," says Bryan.

Reality: It can be--for calls within the campus network.

With POTS, the end-to-end phone connection is kept open for the extent of the entire call. With VoIP, the voice transmission is broken up into multiple packets of information that are coded and sent over the network to the recipient device, which reassembles them.

For calls involving one party that is off-campus--for example, a POTS customer calling from another state--you are at the mercy of the long-distance carriers. In most cases, that's not a problem, because the Baby Bells have expertise in delivering high-quality voice service. But if your POTS service has been bad, VoIP won't improve the connection that goes over someone else's system.

At Clayton State, the quality of the local POTS service had been "awful," says Bryan. It was cumbersome, too. Users had to dial 10 digits just to reach some departments on the same campus.

He reports that one would need excellent or very perceptive hearing to be able to notice problems with his VoIP service. Some of the people in Clayton's music department faculty can sometimes hear a slight difference, for example, but not other people.

Reality: The campus determines the quality of voice service.

The quality of the intra-campus calls, therefore, depend on things such as how much priority voice packets are given over data packets on the network or, if the two run on separate networks, how much bandwidth is allotted to each. If insufficient bandwidth is given to voice, the packets can be compressed to squeeze more information through smaller channels, but that can hurt sound quality. It also requires a system that can assemble the packets quickly, without the delay that is acceptable with assembling an e-mail message, for example, where a delay of 10 seconds between sender and recipient is no problem.

Telephone service is the service that wants to be taken for granted by its users. People expect to pick up the phone and have it work, no questions asked. And that is what typically what POTS has delivered over the past century.

But if a VoIP system isn't set up efficiently, or if compression starts to degrade sound quality, then the comparison with POTS will not go well for VoIP advocates. If there's a 10-second delay in a VoIP phone conversation, one of the callers will have hung up and contacted the IT department to complain.

"A university with a self-contained VoIP network can make the direct connection and it'll be perfect," says Grafton. He tells his higher education customers that if they want to implement VoIP but are not willing to allocate sufficient bandwidth or give voice packets "supreme priority on that network to get to the voice server, ... then it's not going to work."

Big Bend has its voice and data traffic running on the same physical network but using separate "virtual networks," in which the two uses are allocated specific amounts of bandwidth for exclusive use. Beard says voice traffic is also prioritized in case it needs to use more than its set bandwidth. "For example, if everything is running to capacity and voice [traffic] needs 2 percent of what data has, voice will get it," he says.

"The good news is that with most VoIP systems, you as the customer have the option of selecting various quality levels with regard to the voice quality," says Bryan, who notes that Clayton has chosen the highest quality level.

Reality: They can, but they could end up paying much higher communications expenses and supporting an out-of-date campus network.

Voice and data are converging on so many fronts that ignoring it could leave an IHE with much higher communications expenses--and supporting an out-of-date campus network.

Consider the rapid growth in video and other forms of multimedia use on campuses. Online multimedia teaching, research, and collaboration all leverage an institution's voice and data networks, and keeping them separate makes less and less sense.

Reality: Not any more than it is for your critical data systems.

A recent survey of 300 organizations (not necessarily IHEs) carried out by IDC for the Computing Technology Industry Association showed that only 48 percent of the respondents trust the security offered by VoIP. More than three-quarters of the respondents trust POTS security.

And yet, VoIP just takes the voice traffic and runs it over the same network already handling data traffic. Colleges and universities have spent years toughening the security for these systems because that data traffic often includes confidential or proprietary information.

"Security is always going to be an issue, but I don't consider VoIP traffic running within an organization's infrastructure a big concern," says Grafton.

Companies such as NextiraOne have addressed the VoIP security perception by offering services to make sure an organization meets all of the security needs dictated by its own users as well as any pertinent regulations, laws, or compliance requirements.

Reality: True five years ago, but not today.

Tell that to Cisco Systems, which has sold more than 6 million IP phones and counts many universities and government bodies among its customers. But Towle says there were not a lot of VoIP value-added resellers with much experience deploying VoIP in 2003, when Wheelock first launched the technology.

Big Bend's Beard says he loves to be a trailblazer. But he admits that it can be a challenge to be one of the first to adopt a new technology, as he was when he began to research VoIP five years ago. At that time, state guidelines for its colleges required special approval for cutting-edge items, such as VoIP.

But the situation has changed. By the time Beard finished his research three years later, VoIP was no longer a special item on the list and didn't require special dispensation.

Reality: Combined with other network or infrastructure upgrades, VoIP becomes a logical choice.

The decision to switch to VoIP at Clayton State was aided by some mundane real estate factors. "Our administration building is about 50 years old, and it housed our old [voice traffic] Centrix switch," says Bryan. The building was scheduled for renovation, but the architects and engineers reported that it would make more sense to demolish and rebuild the building. So the phone system was going to be relocated anyway, and "the cost to pull copper wires for telephones to a new location just did not make sense," says Bryan. "VoIP was the perfect technology for us."

As part of the switch, Clayton State upgraded its backbone network and switches campuswide to handle more traffic. Similarly, Beard's research at Big Bend showed that a network upgrade and VoIP implementation were best carried out together if they were to work best.

Reality: There are ways to address the problem.

With POTS, the phone company provides the energy for the phone system during emergencies, so users can call emergency services even if there is a power outage. But VoIP runs on a campus network, so if electricity goes out, a VoIP phone becomes a useless paperweight.

There are a couple ways IHEs can handle that problem. Power over Ethernet actually delivers power through the Ethernet system on the campus, providing power to the VoIP phones. (Network switches may need to be upgraded or replaced to handle this functionality.) Also, power generators in the data closets and emergency generators in critical campus locations can help ensure uninterrupted power supplies, notes Bryan.

Beard and Towle both say their institutions retained a relatively small number of POTS phone sprinkled across the campuses, serving as backups.

Reality: VoIP systems generally bundle in many services previously offered separately and at higher cost.

Bryan says that many of Clayton State's departments could not afford voicemail with its former POTS service; instead, they went out and bought answering machines. "We had a hodgepodge of services available to callers when they reached a telephone number," he says. Many of the departments also bought the very cheapest phones available.

With the campus's VoIP system from Avaya, functions such as voicemail, call-center capabilities, and caller ID are included on every phone. Clayton State is also looking at implementing universal messaging, in which voicemail messages, e-mails, and faxes can all be accessed from the same interfaces. It never hurts to be looking toward the next best thing.


Advertisement