Regulatory Clouds Beginning to Clear?

The recently released ARRA NTIA-RUS BTOP-BIP NOFA* (hereinafter simply NOFA) contained two important elements that offer encouraging signs for broadband service providers focused on delivering new, value-added services to their subscribers. To the extent this document expresses the evolving regulatory views of various government organs, these encouraging signs apply even to those operators not planning on dipping into the public trough.

The first is embodied in the term “middle mile.” Just the presence of the term, which is somewhat new in broadband regulatory circles, reveals a deeper understanding into the nature of broadband networks. Most of those dealing with broadband technology have long understood that the “speed” (it’s really capacity, but we’ll stick with the term speed to simplify things) of a broadband service is less a function of the DSL or FTTP access facility than it is of the broadband aggregation network and, for many IOCs, the capacity of their connection to the nearest internet peering point. The NOFA correctly understands that delivering faster broadband is not achievable unless the upstream portion of the infrastructure (the “middle mile”) receives some of the investment. This is true for exactly the same reason that better freeway on ramps will not, by themselves, yield a superior commuting experience.

The existing broadband infrastructure, both telco and cableco, is heavily oversubscribed. Not unlike airlines and hotels, broadband operators sell more bandwidth than they actually have available. This is called oversubscription. But unlike airlines and hotels, which might sell ten or twenty percent more capacity than they have available, broadband operators often sell a few thousand percent more capacity than they possess. Given the bursty and irregular nature of most web traffic, this is precisely the right thing for them to do. It yields affordable services for the subscriber and optimizes capital investment for the provider. But as video becomes more prevalent—and IDC projects streaming video will constitute half of all broadband traffic by 2013—these oversubscription factors will have to be dramatically reduced. And this is where middle-mile investment is required.

The second encouraging aspect of the NOFA is the recognition that broadband infrastructure does more than simply connect subscribers to the internet. The following statement can be found at line 2485 in the NOFA: “Awardees may offer managed services such as telemedicine, public safety communications, and distance learning, which use private connections or virtual private networks, rather than the public Internet.” Although NOFA authors are clearly focused on feel-good public services, this statement implicitly recognizes that broadband pipes (DSL, cable, fiber) can carry a multiplicity of services, only one of which is access to the “public” internet. Others include access to services reachable over “private” networks and walled garden services. In point of fact, the most common “private” networks from which traffic originates are content delivery networks (CDNs). CDNs are networks (and storage facilities) operated by independent corporations (e.g., Akamai, Level 3) that charge for their services. Most, if not all, “internet” video traffic (e.g., Amazon, Apple, Netflix) actually never touches the “public internet” at all; it is delivered by CDNs over private networks directly to broadband operators.

Many network operators have been rightly concerned about this statement within the NOFA at line 2473: awardees shall “Not favor any lawful Internet applications or content over others.” This, of course, deals with neutral treatment of internet traffic, a laudable regulatory goal. But while broadband operators generally concur with these sentiments, they are also seeking new services they can market to their subscribers. Possibly the most desired such service involves the guaranteed delivery of streaming internet video directly to subscriber high-definition televisions.

Technologies are available that allow broadband operators to dynamically detect the presence of streaming video traffic, authenticate subscribers, determine policy, perform connection admission control, and allocate bandwidth to streaming video sessions such that they play back with high-definition fidelity. All that has been stopping the is fear of regulators. That fear may now be somewhat tempered as a result of the foregoing statement embedded in the NOFA.

It would appear that this position clears the way for the following scenario: a broadband operator markets a 10Mbps high-speed internet(HSI) access service over a broadband pipe (ADSL2+ or greater) with capacity of 20Mbps. All HSI traffic is treated in a like manner; nothing is favored, nothing is disfavored. Regulators and net neutrality proponents are happy. The broadband operator also has a connection to a CDN over which Netflix traffic is delivered. For subscribers purchasing a “guaranteed Netflix” service, that traffic is delivered over a separate 1.6Mbps connection (2.6Mbps for HD). The HSI bandwidth slice and the Netflix bandwidth are completely independent of one another. Customers and broadband operators are happy.

The recognition that multiple networks may share the broadband pipe to the subscriber household potentially provides a path forward that satisfies all parties in what has heretofore been a noisy and uncompromising debate. The transformative power of the internet, which many feels requires the balanced and unfettered handling of all internet traffic, is preserved. At the same time however, broadband operators are given the ability to market new services and generate the new revenue streams that are necessary to fund further capital investment in the critical infrastructure.

*“American Recovery and Reinvestment Act National Telecommunications and Information Administration-Rural Utilities Service Broadband Technology Opportunities Program-Broadband Initiatives Program Notice Of Funds Availability.” Only Washington could come up with a 22-letter acronym.