Cable's IP Telephony Conundrum
The Industry's Postponed PacketCable 1.x Push May Prove Too Little, Too Late
APRIL 01, 2004
By Michael Harris, president, Kinetic Strategies, and publisher, Cable Datacom News
The good news is that cable multiple system operators (MSOs) are finally rolling out IP telephony. Large commercial deployments are underway by Cablevision Systems and Time Warner Cable, and other operators like Rogers Cable have unveiled plans to follow suit. The bad news is that a horde of telecom players is also entering the fray, exponentially increasing competition for residential voice services. Compounding problems, the PacketCable 1.x technology that MSOs are belatedly deploying increasingly seems old-fashioned, rather than cutting-edge.
To its credit, in 1995 the cable industry recognized that it had a unique opportunity to be first to market with a best-in-class high-speed data service solution. Willing to take a huge risk for the possibility of a massive reward, MSOs raced to market, initially deploying proprietary cable modem gear to gain early market share. MSOs were so keen on the broadband Internet opportunity that they even decided that sharing the wealth with a partner like Excite@Home was worth it to accelerate deployment. In other words, the cable cowboys decided to shoot first and ask questions later.
Only after launching cable modem service did MSOs optimize their broadband Internet business access models. They implemented standardized DOCSIS technology to drastically reduce equipment and operating costs, and they dropped Excite@Home in favor of their own ISP operations. The payoff has been enormous. North American MSOs now control nearly two-thirds of the residential market and will generate some $9 billion in broadband Internet access revenue this year.
Sadly, with the exception of courageous operators like Cox Communications and the former AT&T Broadband, MSOs have taken the opposite approach on telephony. Instead of shooting from the hip, cable cowboys have simply shot themselves in the foot. Rather than racing to secure a market beachhead with adequate, but not optimized, cable telephony technology, MSOs have fallen victim to analysis paralysis, postponing rollouts until what they perceived to be an affordable solution was ready.
When looking at the big picture, the irony is that the cost savings that MSOs may have gained by waiting for inexpensive PacketCable 1.x gear has been neutralized by the opportunity cost of missing their first-to-market window. Sure, PacketCable MTAs and CMSs are now cheap, but acquiring customers will prove far more expensive when facing off against a plethora of aggressive phone competitors. This is compounded by the prospect that the PacketCable 1.x Network Call Signaling (NCS) platform they're pulling out of the driveway already seems like an IP telephony antique. But honestly, how can this be a surprise when the initial PacketCable 1.0 specification was completed in December 1999?
The crux of the problem for cable is this. Thanks to UNE-P regulations and the entrance of U.S. long distance companies into local telephony, lifeline plain old telephone service (POTS) is already a commodity. And at the other end of the spectrum, the advanced calling features and mobility options offered by Session Initiation Protocol (SIP) players like Vonage blow POTS out of the water for early adopters. PacketCable 1.x implementations fall between these two categories. They usually do not match the reliability of POTS, or even the pricing anymore. Nor do they match the features of IP pure-plays like Vonage. In other words, in IP telephony, PacketCable 1.x essentially offers consumers the worst of both worlds.
On the UNE-P side, major consumer brands like AT&T, Sprint, and MCI are now all selling their own unlimited local and long-distance consumer POTS service bundles for under $50 a month (taxes excluded). These plans, like AT&T One Rate USA, The Neighborhood by MCI and Sprint Complete Sense, often include value-added features like voicemail, caller ID, call waiting and three-way calling. For the low end of the market, they offer even simpler bundles for under $30 per month. One has to wonder: Why would most mainstream consumers risk going to cable IP phone service when they can switch to AT&T for a package that is cost-comparable, more reliable and avoids the installation headaches of a PacketCable MTA? Maybe this is why AT&T now counts more than 4 million local phone subscribers.
And on the IP side, why would techno-savvy broadband-connected consumers opt to buy IP POTS from the cable company and miss out on the cool IP features they can get from players like Vonage, and now even AT&T through its CallVantage service? Vonage's $34.95 per month Premium Unlimited phone plan has goodies like Web- and email-delivered voicemail, free call forwarding and hunting to multiple numbers in any area code, not to mention portability, as customers can plug their phone adapter into any broadband connection in the world and get service on their home number. For its part, Vonage now serves more than 120,000 lines and is adding 15,000 more each month. AT&T says its CallVantage IP phone service will be available in more than 100 U.S. markets by year-end and count at least 1 million subscribers by the close of 2005. Even RBOCs Qwest and Verizon are readying their own IP phone rollouts.
All is not lost for cable, of course. If cable operators play their cards right, they could offer the best of both worlds and deliver a truly differentiated consumer IP telephony offering. The answer for MSOs is PacketCable Multimedia (PCMM). Using the architecture outlined in CableLabs' PCMM specification, it is possible for MSOs to guarantee quality of service (QoS) via DOCSIS 1.1 or 2.0 cable modem connections for virtually any IP multimedia application, including SIP services. To understand how PCMM and SIP work together, read on ("Combining PacketCable Multimedia And SIP"). |
