September 1999

The Age Of Convergence: Threat Or Opportunity?

BY IKE ELLIOTT

The big communications service providers must be worried. As the packet telephony revolution continues to pick up steam, and the costs for network equipment plummet, the incumbent communications service providers must not be sleeping well. After all, new carriers will soon invade their markets, spending a fraction of the capital incumbent carriers deployed 10 years ago to build smaller networks. Won’t the new entrants have a competitive advantage?

On the other hand, can’t incumbent carriers deploy the new converged network equipment, too? Of course they can, and they are busy readying their own convergence battle plans. Just look at AT&T’s cable purchases and Sprint’s ION strategy to see two clear examples of incumbent carriers gearing up for a convergence fight. Both companies are embracing new converged network architectures, further validating the move toward packet network backbones and away from circuit switching technology.

While the new technology will increase competition and create opportunities for new players, the threat to the incumbent carriers is softened by the fact that service providers of all sizes and ages can share in the benefits of the new architectures. However, incumbent service providers risk falling into a trap if executives are unwilling to cannibalize their own telephone revenues in order to move aggressively into Internet telephony.

THE COST FACTOR
Service providers all over the world are looking into packet telephony as a cost reduction mechanism. While some focus on ATM, and others on IP, there is widespread agreement that equipment using either technology undercuts the cost of circuit switching in terms of bits transported per dollar. Plus, distributed software paired with packet equipment can now emulate the functionality of a class 4 or class 5 switch, so the functionality lead once held by circuit switching is beginning to erode. The industry has begun to refer to distributed switch emulation software systems as softswitches.

Currently, a distributed softswitch costs 30 to 70 percent less than a circuit switch with equivalent capacity. While that figure is impressive by itself, the performance/cost ratio for softswitches doubles every 20 months, as compared with 80 months for circuit switches. Softswitches have a cost lead, and are poised to run up the score. These cost trends exist without regard to any temporary regulatory arbitrage phenomenon, which may have cost justified Internet telephony equipment in the early years. Service providers ignore these trends at their peril.

UNCHARTED TERRITORY: APPLICATIONS
Even if service providers could not count on the coming cost advantages of Internet telephony, carriers must still adopt the new architecture or risk failure. Why? Because of the promise of new packet-based applications that will entice users of all forms of interactive multimedia communications. Service providers using only standard telephones and circuit switches cannot deliver many of these applications. Some examples: High-fidelity audio quality, voice/video integration, voice/data conferencing, stereo conferencing, and wireless PDA/voice integration.
And the point of the coming application wave isn’t really that we will be able to do things that the old circuit-switched world could not support. The point is that communications network users will have the choice of which applications they would like to run.

The CTI market has tapped the choice theme for years, and to great advantage. CTI enabled end users to create their own services by purchasing software/hardware packages that delivered the precise functionality needed by that end user. When contrasted with the tired and overly complex advanced intelligent network (AIN) service creation concepts preferred by some carriers, it’s no wonder most network users abandoned AIN to focus on CTI instead.

The powerful drive for choice will now cause customers to select Internet telephony service providers over circuit-switched service providers. Successful Internet telephony service providers will expose CTI-like control interfaces on their IP multimedia networks on top of softswitches. These new control interfaces will allow call routing, media processing, and call treatment from an independent, third-party application. End users will purchase software, which delivers the precise functionality they need, but now the software will run on general-purpose computers with IP interfaces, instead of special purpose hardware with circuit interfaces. Network users will not only be able to choose the solutions they want, but they will also be able to choose where the solution runs — on their own enterprise server, or hosted on a server operated by their service provider.

In effect, service providers will begin to position their network as a network operating system, upon which many applications may execute. As it was in the early days of Microsoft, the network operating system with the most applications will begin to dominate.

CANNIBALIZE OR DIE?
Many large service providers have built value-added services businesses in an attempt to preserve margins in a competitive carrier world. Examples include 800 services, virtual private voice networks, calling card services, operator services, call center services, fax, and messaging services. Much of the value these carriers have added never quite meets the true needs of their most demanding customers. Something inevitably is lost in the translation between customer requirements and service development engineers, and the resulting product often falls short of customer expectations.

Despite the error-prone product development cycle of service pro-viders, very large multi-billion dollar businesses have been built in several market segments, and most incumbent carriers are reaping the benefits of their product development efforts in enhanced services.

Against this backdrop, the network operating system is beginning to take root, allowing customers to meet their precise needs by purchasing a more primitive “operating system” interface from their service provider, and then purchasing a third-party application independently. What does it mean for the established enhanced services offerings of the incumbent carriers?

THE CRITIC’S PICKS
The development of network operating systems will be one of the most interesting spaces to watch in the coming years. Some carriers may be reluctant to give up their enhanced services revenue streams, and so they will either offer no network operating system product, or will price their network operating system product so that it does not cannibalize their existing revenues. Other carriers may dive in despite the risks. Also, it will be fascinating to watch whether the network operating system model takes root quickly, or only gradually takes market share.

Meanwhile, watch the carriers that are first to market with carrier-grade Internet or packet telephony services, and especially those carriers that have demonstrated operational readiness to support rapid growth in Internet telephony subscribers. These carriers will be the ones to grab the lion’s share of the new Internet telephony market.

Just as importantly, watch the carriers that don’t execute a complete convergence strategy, including both broadband and packet telephony components. These carriers are likely to lose market share in the coming years.

Ike Elliott is Vice President of Softswitch Services at Level 3 Com-munications, Inc. Level 3 is a communications and information services company that is building an international advanced IP technology-based network. This network, consisting of both local and long-distance networks, is expected to be completed in phases over four to six years. Level 3 will focus primarily on the business market, using its IP-based network to provide a full range of communications services — including local, long distance, and data transmission as well as other enhanced services. For more information, visit Level 3’s Web site at www.level3.com.