How the IP Telephony World Unfolds

Opinion: After a week at Supercomm, the big telecom show, some reactions. Yes, IP telephony will pervade the enterprise, and no, it probably won't be the startups who win the voice carrier spoils.

After spending four days among the true believers of IP telephony—virtually every vendor who spoke at Supercomm—I can envision a world in which telephone service over dial-up lines becomes as clear a sign of technical backwardness as dial-up Internet connectivity is today.

Voice over broadband data networks could become that pervasive. Market research says IP phones in the enterprise wont outnumber traditional TDM phones until 2009 or so, and that sounds right, but here are some of the signs and seers pointing in that direction at last weeks show in Chicago:

Vendor push to service providers: Many of the core exhibitors at Supercomm—those like Cisco Systems, Siemens Information and Communication Networks, Nortel Networks, Lucent Technologies and Alcatel—devoted the bulk of their booths to the shows primary market, which consists of service providers of various stripes: telecom carriers, cable companies and Internet service providers, and various combinations thereof. These exhibitors produce the core switches that carriers have long installed in their central offices to terminate subscriber lines, aggregate them into long-distance trunks and route calls. What they show the SPs now is very much about provisioning IP Centrex.

Vendor push to enterprises: These same exhibitors (except for Lucent, whose enterprise arm got detached and was renamed Avaya) are also the manufacturers of enterprise PBXs. These are essentially smaller versions of carrier switches. They terminate desktop phones, switch between them and pool the use of outgoing trunks to the wider world. These too are all being configured, at least optionally, for IP call transport and the multimedia, multisite communications enabled thereby.

To the enterprise customer, these PBXs are promoted as (among other things) new instruments of self-sufficiency, ridding companies of the need to call in telecom field service personnel every time new extensions must be added or changed. To the service provider, the manufacturers pitch the IP PBX as just one option to be offered to their customers on a continuum of self-sufficiency, ranging from pure IP Centrex (carrier-switched, customer only buys phones and service) to hosted (customers own PBX is located at and managed by the carrier) to managed (customers own PBX is located on the customer premises but is managed remotely by the carrier).

The vendors stress the consistency in end-user experience between carrier and enterprise versions of their platforms. I take this to mean that if I, as IT manager, decide to test the IP Centrex waters by tying in one branch offices worth of IP phones and then later buy my own softswitch and media gateway platform, I need not replace the phones or retrain my users.

Low barriers to entry for new VOIP carriers: "I worked for a CLEC [competitive local exchange carrier] startup years ago, and it was $30 million to wire up every city," said Denise Grey, chief marketing officer of Verso Technologies, in Atlanta. "You can buy a softswitch and your first application—say prepaid calling card long distance—from us, and get there for half a million dollars now." A long-distance resale arrangement with a wholesale carrier who has gateways in every major market—someone like Level 3 Communications—buys a network with reach to all non-IP phones.

Next Page: Why incumbent telcos can keep their local lines all to themselves.