By Matthew Friedman via Networking Pipeline
The promise of voice over IP (VoIP) is hard to resist: a single network carrying data, voice and other traffic, enabling a wide range of advanced communications applications. And all this can be implemented with comparative ease and with substantial cost savings over traditional telephony.
It's too bad that it's not completely true.
Make no mistake, VoIP is a big part of the future of voice communications, and its benefits are legion. The problem is that the real advantages of the technology are often obscured by a combination of genuine excitement, optimizing, unrealistic expectations and marketing hype. VoIP is the future, but it's a future shrouded in myth.
"The whole point is that people are moving in that direction," says Forrester Research vice president Elizabeth Herrell. "But I wouldn't want people to go into the future blind."
One of the most persistent myths about enterprise VoIP is that, right out of the box, it's a money-saving technology. The truth, however, is a bit more complicated, according to Herrell. "Sometimes it is, and sometimes it isn't," she says. "It's really situation-dependent -- in some cases, enterprises have saved a huge amount, and then some haven't."
The promise of deep cost savings has been around since the earliest days of VoIP, when it made the jump from technoweenie hobby to prime-time enterprise technology. "There was a lot of talk then, and even now, about how much money you could save," Herrell says. "But the reality is that the advantages of VoIP aren't in cost savings. What is gives you is new applications and future-proofing."
This is compounded by the belief that the cost of deploying VoIP rests primarily with the hardware. Admittedly, the technology can require a substantial capital investment, but that's only the start. Plugging voice into a network that isn't ready to support it is, at the end of the day, an expensive waste of time.
"The myth is that the highest cost of the network is the network itself," Herrell says. "The truth is that it's the upgrade, and the upgrade for VoIP is a massive issue. It's something that enterprises have to consider at the outset, but many don't."
And the technology itself isn't out-of-the-box easy, either. Herrell points out that enterprises sometime stumble into VoIP with the mistaken belief that all they need to do is deploy the systems and everything will be okay. Infonetics directing analyst Kevin Mitchell agrees. One of the biggest VoIP myths today is "that it's easy, plug and play," he says. There is "still lots of interop work, software tweaks that need to be done by vendors to get media server from vendor A to work with softswitch from vendor B and the app server from vendor B."
Indeed, interoperability is another issue often forgotten in the excitement around VoIP. "The myth is that VoIP is completely open, but the truth is that it's still a proprietary technology," Herrell says. "You buy into a total package."
Part of the problem is ensuring sufficient quality of service to support dynamic, highly sensitive voice traffic. While it's true that most call servers are network agnostic, and consequently, your underlying, basic network gear will work perfectly well with most VoIP systems, switches and handsets are a different matter. "Even though the applications run on open source servers, vendors have written proprietary extensions," Herrell says. "Cisco's AVVID [Architecture for Voice, Video and Integrated Data] is a perfect example of that."
Moreover, even the openness of the underlying network can be problematic and, according to Mitchell, interoperability requires a considerable amout of work --work that service providers are sometime loathe to take on. Service providers "don't want to be system integrators and deal with managing all those vendors," he says.
To some extent, the Session Initiation Protocol (SIP) is supposed to solve some of the vendor interoperability issues, but Herrell says that the idea that SIP will usher in a golden age when switches and handsets from different vendors will work together in perfect harmony is, in fact, a myth. "SIP handsets will interoperate, but they won't offer all features when they do," Herrell says. "Due to the immaturity of the standard today, vendors will enhance their products with all kinds of proprietary features."
Ultimately, the biggest myth of all -- and the foundation of all the other myths -- is that voice is just another kind of data. It's a myth spun, according to Herrelll, by equipment vendors who want customers to buy into VoIP as just another network application. "The big myth for a long time has been the idea that you'll have a single network cable, and we've seen that this just isn't the case with data, storage and with VoIP," she says. "Voice is not just a LAN application, it's a discrete technology that requires special expertise and considerations. It's not just another data application."
The bottom line is that replacing a telephone system within or without the enterprise is an immense undertaking. Most enterprises have had their current private branch exchanges (PBXs) for decades and moving to a new communications paradigm is considerably more complex an nuanced than plugging in new hardware. "It's important that companies understand this," Herrell says. "They need to have the right organizational skills and planning to make it work."
