The DSL Gateway Dilemma
For many, the answer is, "More than we were before." That's why a growing band of telcos have joined the Home Gateway Initiative (HGI) to create a de facto standard RFP.
But how much more, exactly? As discussed in our latest Heavy Reading report, DSL Gateways: Beyond the Router, there are many competing views, and many open questions in an area that is both strategically critical and potentially treacherous.
For telcos and those using DSL, the renewed interest in the broadband device in the home, and what runs through it, is understandable. Most major broadband service providers now want to provide packages of services that include, at a minimum, VOIP, and prospectively a range of entertainment and other services as well.
To do so, they need a manageable environment in which services can be automatically configured, updated, and quality-controlled, to minimize customer service calls. And that implies a gateway that does, and is, a lot more than the conventional DSL router or modem that most customers still have. So far, so uncontroversial. But questions about this transition abound.
First, what should we include or specify in the gateway itself? Almost all service provider gateways now include one or two FXS telephony ports for VOIP, although there are still differences of opinion over which VOIP software stacks to support. But should we include functionality normally found in set-top boxes, or add software and ports to support services such as home security and monitoring? There are only a few gateway vendors and service providers moving forward here, yet given the steady evolution of the gateway from the early days of simple one-port DSL modems, there seems little doubt that complexity and functionality can only continue to increase.
Second, how far do we allow customers to configure services and attach devices? Some service providers and a few vendors argue for a device and an environment analogous to that which exists in cable MSO networks, where customers have little or no ability to control the cable box. But over the past five years, broadband users have become increasingly sophisticated and, in effect, emancipated, buying wireless routers and other home devices that are not supplied by the service provider, and running a range of services over them. Are they really ready to relinquish that independence to the telcos?
Third, how far into the home are we prepared to go? From the customer's point of view, if the service provider is supplying a branded box to deliver a package of services, then it is also responsible for everything that happens beyond the gateway. But that represents a huge and potentially risky extension of the operator's commitment to the customer, with potentially disastrous consequences. There is a great deal of proprietary technology in the home, and little agreement on some crucial technologies.
For example, should we support only wireless home networks? And if so, do we go with a proprietary enhancement to 802.11 to improve throughput, range, and QOS, or wait for the much-anticipated 802.11n standard? And what about the new wireline home network standards, such as HomePlug AV, MoCA, and HomePNA? Which, if any, should we support? Will they help or hinder our relationship with customers? There are no easy answers to these and related questions, yet they cannot be ignored or postponed if service providers want the gateway to take them right into customers' homes.
Finally, how do we manage the gateway effectively? Better management of broadband connections and services is the core justification for service provider involvement in specifying gateways, but despite the successful completion of the Broadband Forum 's TR-069 standard, much remains to be done to achieve interoperability and to manage effectively beyond the router, and there are many proprietary fixes to choose from in the meantime.
It may be that there is no halfway house here: Telcos must either go all the way – as Orange (NYSE: FTE) and Telecom Italia (TIM) have done, investing heavily in customized gateways and related home devices in order to tie in the customer and boost ARPU – or take a much looser approach, in which many or most customers continue to buy the gateway from a retail or e-tail outlet – albeit a gateway that is ready for VOIP and other services, and is even tied to a specific service provider's offering.
Just how far the big broadband service providers will drive the continuing gateway evolution remains an open question, and providers will need to think through the business case very carefully as they move forward. Most likely there will be differences in approach. But – to end on a positive note – there seems little doubt that more and more functionality will appear in gateways, making it easier for customers, telcos, and third-party service providers to add value to broadband connections. With some 40 vendors now vying for attention, there is no doubt that this dynamic sector will continue to help drive broadband speeds, services, and penetration for some years to come.
— Graham Finnie, Senior Analyst, Heavy Reading