I thought it was a very good piece. with excellent clarity. It defines companies as falling into two groups. Plan ‘A’ and Plan ‘B’. The first are companies focussed on interoperability with the phone system. The second are pure computer-to-computer players.

Santa Cruz Networks (my new company) is very definitely in the ‘Plan B’ space. We provide the network devices necessary to enable a service provider to offer voice video and data communications, to individuals or groups, over IP. Our platform is a true many-to-many platform. Edge devices need no more than 100k of bandwidth, no matter how many people are in a meeting. And all end points have full voice video and data communications to all other end points.

This whole era and its process reminds me a lot of when I started Easynet with David Rowe in the UK in 1994. At that time BT were not offering internet service. We were No. 2 into the market. By 1999 there were 2000 internet service providers. UUNet and Earthlink and MindSpring and Spry all came out of that era. The Telco’s were late to market and tried to slow it down. Many tried to build walled garden, AOL like, hybrids. In the end their hybrid strategy failed. Ascend made a multi billion dollar business supplying the emergent ISP’s, as did many others.

The similarity is that now, ‘Plan B’ is available at a price point that allows new communication service providers (CSP’s lets say), to enter the market with pure IP based, computer-to-computer services that bypass the phone system entirely. Think desk-to-desk or home-to-home or desk-home calls.

My bet is that we are going to see lots of CSP’s arising over the next few years.

We have sold our 2000-subscriber , multi-party, Voice Video and Data over IP, appliance to many new service providers since December. Their price is only $85-100k MSRP, depending on which model, to put up to 2000 subscribers into service. Discounts mean that in pracctice this is even lower. Bandwidth is about $100 per month per 50 subscribers. That makes for a great business at very low price points. And the performance [great voice quality, application and document sharing, video, IM and presence, up to 200 on a call) is mind-blowing.

I think there is a lot to talk about in the edge and center issues here also. P2P and client-server both have an interesting place in IP based communications. It’s very clear that you need both, depending on what service and what QoS you want to provide client-server is actually the only way. At simpler service levels (1-to-1 and 3-way calling) P2P works quite well.

Another great issue is interoperability within ‘Plan B’ approaches. We have decided to standardize on email address as the ‘telephone number’ for IP based communications. It seems easier to explain an upgraded email address that to create a whole new ID system. Skype has a ‘screen-name’ approach. Beyond identifiers, there are really no protocols in the standards space for multi-party voice, video and data calls. Hence we are all proprietary. How are we going to get beyond that?

Anyway – I enjoyed Clay’s essay and look forward to how this area evolves.