Everybody talks about digital convergence, but very few companies actually deliver on that promise. After serving in top executive positions over the last decade at Lotus and Shiva, Woody Benson is now president and CEO of MCK Communications, which specialises in making equipment that extends the network functions of a PBX across an IP network. Benson recently talked to Michael Vizard about what IT organisations can do to make use of IP today to extend their phone systems by marrying digital and analog technologies.
IDG: What trends are coming together to push convergence in general, and products such as yours?
Benson: All of a sudden, broadband is hip and happening. On the cable side it's obviously getting well funded by AT&T, which is injecting $US80 billion into it. There's this new, natural, excellent network to deploy our stuff across, unlike ever before. Our notion is to be the company that enables the broadband people and the service providers to extend enterprise applications to individuals in branch offices, on really any kind of an end-to-end VPN [virtual private network].
How does it work?
We do remote PBX extension on analog and ISDN lines and leased lines. In the last year, we also came up with a really interesting product, called the Branch Office EXTender, which is a multi-user version of our product for remote offices. It essentially can be the piece to [attach a] client to a company's data backbone over a wide area network interface. We extend the full functionality of the voice switch, which includes all the features of the PBX, the software on the PBX, call accounting, the numbering plan, etc. A gateway device goes onto the PBX - we call that the PBX Gateway IP. And then there are the remote units; we have a number of different remote units going into individual homes or branch offices.
What is the benefit of doing that?
PBXs aren't going out of Fortune 5000 companies for a long time yet, but there are lots of solutions for data and very few solutions for voice. The big voice players have great solutions for about 50 people. But for one to 50 people, there's no real alternative. We're the company that provides the extension to the installed base of voice systems cross-company. We call that extending the virtual cubicle.
So the basic idea is that if customers call in, they should not be aware of when they are talking to somebody in a remote office or even at home.
Exactly. Our philosophy is there's lots of capacity in these big switches, and the big switches are only getting better.
In the struggle between data and telecom, who is winning?
Telecom guys clearly are losing the long-term battle and they know it. But they are the lifeblood of the organisation, and they are deep-rooted in high reliability. I think the data guys have also underestimated the size of the telecom hill. I think the data guys are a little ambitious and haven't really quite scoped the size of the hill. It was really interesting. Voice revenue is way ahead of data revenue. And there's nothing more precious in an organisation than dial tone. It's going to be a long time before there's one IP network that does that.
Is it cable or DSL that will drive convergence?
Broadband is going to go from basically a million or two million today to 30, 40 million in five to six years. I think there's a distinct split between DSL and cable. I firmly believe that DSL is going to go to the business community, and cable is going to go to the consumer community and that there's going to be DSL service providers whose essence of business is selling back to enterprise accounts. People already have prototype 28Mbps designs on DSL.
What are the challenges associated with sending voice over IP?
It's a data highway, and there's a voice lane being formed. It's really up to the Ciscos of the world - and [to] people getting smart about adhering to their quality of service [QOS] algorithms. Essentially, the problem isn't that difficult to solve. We can get voice packets as small as 10K, so over a data pipe that's not asking for a lot of bandwidth to be carved off. What we're really asking the QOS guys to do is make sure we always get that 10K in case any fat packets that come through want to take up the entire bandwidth. I think you're also seeing a real proliferation of ATM, which is very helpful to voice applications. My guess is that WAN QOS will happen way before LAN QOS because WAN QOS will be managed by 20 guys and LAN QOS will be managed by 300,000 guys. That's going to be a little more challenging. Hopefully, Cisco will drive this. I don't think there's any other win for anyone else having a different QOS scheme. After all, how many QOS schemes can a network have?
So how will all this come together in the end?
It's a great geopolitical situation where you have Lucent saying make your data network as reliable as your voice network. And then you have Cisco saying voice is free, make it part of your data network. What's really interesting is that so much energy is being put towards carriers and big IP switches that people have neglected the enterprise in a lot of this movement. I think you'll see a major resurgence coming back towards the enterprise.
As a company, how is MCK doing these days?
We're a 90-person company. We just finished our fiscal year at the end of April, and we did close to $20 million in revenue. We're actually profitable and cash-flow positive and grew 80 per cent year over year, which is all kind of rare in this day and age.
How did your interest in remote access at Shiva lead to this stint with MCK?
After I left Shiva, it hit me like a ton of bricks. We did all this remote access stuff, and we forgot the phones. We forgot voice in the whole equation. I can't believe we just simply forgot. We talked about remote access, but it's just the data part. So in MCK, I saw a company with the opportunity to take those PBX extensions, which is really extending your phone and everything that goes with it, over data networks versus using just circuit networks.