Let's see, when did all the hype about convergence begin? I dug up a Network World story from 1997 that described how convergence was the talk of Networld + Interop (N+I).
Here we are four years later, and I think it's fair to say that convergence hasn't made much of a dent in the corporate network.
In fact, Gartner estimates that IP/PABX shipments constitute less than two per cent of all PABX shipments, based on an analysis of the market in the fourth quarter of last year.
And when Miercom asked nearly 100 voice-over-IP vendors to estimate what percentage of voice traffic is travelling over an IP network at least part of the way, the vendor estimate averaged out to nearly 8 per cent.
However, the vendors also estimated that the installed base for voice-over-IP equipment is more than 70 per cent service provider and less than 30 per cent enterprise, which puts enterprise-based voice-over-IP traffic in the 2 per cent range.
So why hasn't convergence taken off?
n According to Gartner, voice over IP was never something that customers asked for. It was a vendor idea pushed by data networking/IP-centric vendors like 3Com, Cabletron Systems and Cisco Systems to attack the PABX vendors.
- Until very recently, vendor products didn't scale far beyond 75 end users.
- Standards are still a mess. There's H.323, there's SIP, there's MGCP. Gartner predicts that standards won't be ironed out until 2003.
- The return on investment for voice over IP is difficult to prove.
- Voice-over-IP networks still can't beat traditional voice for breadth of features, security and reliability.
- New applications - unified messaging, Web-enabled call centres - have yet to gain traction.
- Interoperability remains a major concern.
- And quality of service was listed in the Miercom survey as an important factor in the slow rollout of voice over IP.
I visited Cisco earlier this year and was very impressed with the company's voice-over-IP network, which sports all the bells and whistles - unified messaging, voice activation, browser-equipped IP phones. So Cisco is indeed eating its own dog food. The problem seems to be that other companies aren't biting.