BROADBAND COMMUNITIES is the leading source of information on digital and broadband technologies for buildings and communities. Our editorial aims to accelerate the deployment of Fiber-To-The-Home and Fiber-To-The-Premises.
Issue link: http://bbcmag.epubxp.com/i/325023
6 | BROADBAND COMMUNITIES | www.broadbandcommunities.com | MAY/JUNE 2014 BANDWIDTH HAWK F ast lane. Slow lane. Net neutrality. Megamergers. What, if anything, should people worry about? We at BroadBand Communities tend to focus on one thing: bringing fber-enabled ultrabroadband to as many people as possible. Reliable, ubiquitous bandwidth is important for the U.S. economy, lifestyle and well-being. An ever- increasing proportion of people's activities is broadband-related – whether entertainment, shopping, health care or education. What's the best way to get people the bandwidth they want at a price that does not crowd out other economic activity? Broadband deployments are expensive, no matter what the technology. Interest rates are beginning to rise. How would smaller players such as Time Warner Cable and DIRECTV – each with about $20 billion in debt – raise funds for expansion when investors believe their debt is already too high? Large owners of business and residential property already complain about smaller providers' not having the resources to improve service. Is there a slow lane on Internet-connected private networks? Tere certainly is. Netfix, which right now is responsible for nearly one-third of all North American downstream Internet trafc – and, by some calculations, more – enjoyed average access speeds to Comcast consumers at a fairly steady 2 Mbps all last year. Tat speed dropped to 1.4 Mbps in January, and within a few months jumped to 2.6 Mbps after Netfix agreed to pay a toll. Is that fair? Most economists will say, in guarded terms, that it is. Any users that consume enough of a scarce resource that the providers of that resource have to spend extra themselves, should help pay for the extra capacity. But this picture sufers from jitter. Netfix is today's bandwidth hog. But Netfix also has long ofered to place its own servers deep in providers' networks to ease trafc congestion. And within the past decade, BitTorrent, YouTube, other Google divisions, and Facebook have all taken a run at the top of the nationwide bandwidth consumption list. Who's next? How could the FCC regulate? Tough proponents of net neutrality have raised the specter of innovative startups' collapsing under the weight of fast-lane tolls, startups are by defnition small. Te limit for megabytes moved can be set by the FCC at a fairly high level, before which the fast lane would be free. But what about regional data centers, which together form that "cloud thing" everyone is talking about? Does a regional data center run by Comcast or Google or Amazon or Microsoft get away without nationwide tolls? Without local tolls? By no-cost peering? By peering with cost diferentials? A network is only as fast as its slowest pinch point. Te Internet, with multiple meshed switches, bypasses pinch points to some extent, but congestion has gotten worse. Te FCC would rely on content providers to complain. How would tolls be priced? Te per-meg cost of transport has been falling. How does that square with the supposed scarcity of bandwidth? Simple: Congestion now is at the consumer end, the network edge. Tere isn't enough fber there. But consumers and businesses are uploading more and more to the cloud. Tat's not historically been an edge activity. What else do the FCC and bankers have to deal with? Consumers prefer to buy video à la carte – not by the channel but by the specifc event or movie or series. Te biggest carriers currently have access to content at a lower cost than smaller carriers, but that advantage erodes with the growth of à la carte – and the erosion scares of investment. What should be done? Legislation to allow the FCC to structurally separate network operators from service providers would vastly increase revenue potential for network builders and guarantee that loans get repaid and capital earns a good return. It would stimulate network investments. It would also stimulate innovation by making any content or service quickly available nationwide. Every lane would soon be a fber fast lane. Every carrier would have nationwide reach. What are the chances of that happening in the United States? Absolutely zero. We'd much rather add legal and regulatory uncertainty to poor technology. v Contact the Bandwidth Hawk at email@example.com. Every Lane Should Be A Fiber Fast Lane The FCC needs new legislation to regulate the Internet's merging fast lanes. That doesn't appear politically possible. By Steven S. Ross / Broadband Communities BBC_May14.indd 6 5/29/14 9:14 AM