It’s clear that telecommunications networks are getting faster. An important question, which is the subtext of this interesting article by Jon Brodkin at Ars Technica, is just how fast these networks have to be.
The world of technology moves quickly. A few years ago, carriers, mostly cable operators and telcos, with a few others types thrown into the mix, struggled to provide subscribers with enough bandwidth to keep their services running smoothly.
That still is the case in many areas. But, increasingly, broadband services as fast as 1 gigabit per second (Gbps) are available and no longer are too much of a technical challenge. Thus, the question becomes: Where does it make sense to build the advanced platforms that approach the 1 Gbps level?
Brodkin’s piece juxtaposes the cable industry’s ability to reach 1 Gbps with its willingness to do so. The industry long has been able to reach the speed threshold through the widely deployed third version of the Data over Cable Service Interface Specification (DOCSIS 3.0). Even more efficient ways of hitting the 1 Gbps mark will be available in DOCSIS 3.1, the fast-tracked next version of the standard.
The reality, according to the story, is that the industry largely is not choosing to offer those speeds to its residential customers. Bodkin sets it up by first suggesting that the cable industry has been claiming superfast services for years:
Cable companies haven't been ignoring this consumer demand… but they haven't done anything to satisfy it, either. Comcast demonstrated the "first ever 1Gbps broadband speed download over a production HFC [hybrid fiber-coaxial] network" two and a half years ago at the NCTA [National Cable and Telecommunications] conference, and the company showed off a 3Gbps technology at this year's cable show.
He concludes with the reality that the multiple system operators (MSOs) largely haven’t made those platforms available:
It is almost a reflex to blame cable companies when confronted with seeming contradictions. But the reality appears to be a bit less dramatic. The cable industry isn’t rolling out these plans because sufficient demand doesn’t exist.
Cable operators are business people and their industry is in a state of unprecedented transition in the services it offers, the devices those services are sent to, and the formats and standards that are used for those transmissions. The last thing that they have the time and money to do is ratchet up the speeds to a level that is required by a small minority of subscribers. Demonstrations at shows are exactly that: demonstrations of what is possible. They don’t constitute commitments or promises.
The issue becomes even more interesting when considered in the context of Google Fiber. The company has garnered endless headlines during the past couple of years with its high-speed network. The question becomes whether the usage patterns from the earliest systems may take some of the bloom off the 1 Gbps rose.
A story at KansasCity.com focuses on holes in coverage of the Kansas City metropolitan area upon which Google Fiber has to a great degree concentrated since it launched. Though the story touches only briefly on the financial aspect, it is fair to wonder whether at least a part of the discussion of how quickly to proceed is demand.
But work goes on. In late November, Prairie Business, a site in Grand Forks, North Dakota, reported on the Dakota Fiber Initiative, a public/private partnership aimed at vastly upgrading North Dakota’s Internet speed and bandwidth. The consortium, which plans a pilot project in Fargo, sees 1 Gbps services as the initial speed target, with faster service to come later.
It is clearly too early to suggest that carriers are overestimating demand for fast services. Indeed, the relative slowness of cable operators to offer full 1 Gbps services to residential customers may be purely a matter of definitions: Cable operators offer such services, but aim them at small- and medium-sized businesses through their commercial services arms.
Google Fiber and other providers simply may aim to concentrate all users short of huge enterprises on a single network served by unified technical, sales and marketing staffs. If so, it will be interesting to see which model – one in which SMBs are served separately and one in which they are lumped together – is more successful.