The Economist: Why LTE can't substitute for fiber

Some believe the Internet will become untethered over the last mile and point to cutting edge wireless transmission technology known as LTE or 4G. The two biggest telcos in the U.S., Verizon and AT&T, are rolling it out (or are about to in the case of AT&T.)

But it won't be able to replace the nation's aging copper cable infrastructure that has grown increasingly difficult and costly to operate reliably. Nor is it likely to provide sufficient capacity for future growth in bandwidth demand -- something that Verizon and AT&T are acutely aware having faced growth pains and capacity constraints with their current generation of 3G wireless.

The Economist explains why:

Already LTE has shown itself good for at least 5Mbps—impressive for a mobile technology still in its infancy (see “Generational change”, December 3rd, 2010). But with peak speeds of 1Gbps theoretically possible, LTE’s next iteration should make downloads of 100Mbps over the airwaves a matter of routine. Developments beyond that could lead to near-gigabit speeds.

Of the two, though, a fixed link like fibre remains the better bet. Sooner or later, even a 4G wireless protocol such as LTE or its country-cousin WiMAX will become overwhelmed by the exponential growth of mobile traffic. By contrast, an optical link to the home could use a multitude of different wavelengths to boost throughput almost indefinitely.

Network World also weighs in:

So the next question about wireless broadband as a substitute. Recall that according to the U.S. Government Centers for Disease Control and Prevention (CDC), 1 in 4 homes has cut the legacy wireline phone cord in favor of wireless-only voice. Could we see wireless substitution rates that high for broadband access? We think not because radio spectrum is a limited resource, and unlike wireless voice networks that have plenty of spectrum to manage voice calls, if 25% of broadband users shifted from wireline access, the demand for wireless broadband would likely exceed available spectrum given today's technology.

The post-broadband era begins

As 2010 draws to a close, we are also seeing the closure of a chapter of the early Internet era and the beginning of a new one. The first chapter opened in the early 1990s when the few people who connected to the Internet did so with narrowband "dial up" connections using the Public Switched Telephone Network (PSTN). By the end of that decade, dialup evolved from 1200 and 2400 baud connections to 56Kbs connections as well as ISDN offering Internet connections of up to 128Kbs. At the same time, "broadband" began emerging with DSL and Internet services offered by cable companies.

We are now beginning a new chapter where throughput speeds that defined an Internet connection will be less relevant than the services and applications people use when they access the Internet. If the connection can't support them, it no longer will be considered bona fide Internet-based service. From a practical standpoint, that means dialup and satellite connections are now obsolete since they cannot provide end users a full Internet experience due to the inherent physical limitations of their technologies.

Also being rendered obsolete as bandwidth demand grows exponentially, particularly with the explosion of video content and mobile Internet:

-- The U.S. Federal Communications Commission's definition of a "broadband" Internet connection as 4Mbs down and 1Mbs up.

-- The term "broadband black hole" and dubious efforts to "map" these locations. These areas will simply be regarded as disconnected from the Internet, similar to the "off the grid" term applied to those locations lacking electric power service.

FCC ruling heralds regulatory end of "broadband" era

The U.S. Federal Communications Commission's decision this week barring providers of fixed premises wireline Internet connectivity from selectively rationing (or blocking) bandwidth based on the source and/or content marks the regulatory beginning of the end of the "broadband" era. Whereas broadband once defined a premium telecommunications service offered at a premium price, the FCC is effectively declaring that an Internet connection is an Internet connection regardless of what information or content consumers receive from it. ISPs cannot devote greater bandwidth to some information or content (for example, their own proprietary content) while affording less bandwidth to other information or content.

This is the right decision that recognizes the Internet as a de facto common carrier telecommunications network similar in that regard to traditional voice telephone service. The large national legacy telcos and cable companies warned before the FCC adopted the ruling that it would discourage capital investment in their infrastructures, which the FCC noted early this year continue to leave an estimated 7 million U.S homes offline.

There's no evidence the business case for more network investment by the large national legacy telcos and cablecos would have been improved had the commission come down on the other side of the issue. Their business models are constrained by the need to pay their shareholders generous dividends as they have done for decades and by high labor costs to modernize and build out their plants outside of densely populated urban and suburban areas.

NTIA report reinforces outdated notion of "broadband adoption"

The National Telecommunications and Information Administration (NTIA) is issuing a report today that continues to promote the outdated notion that Internet connectivity is separate and distinct from other types of Internet delivered telecommunications such as voice and video. It does so by parsing out "broadband" usage among various demographic groups.

Unfortunately, it's about as useful as reporting distinctions among these groups in their landline long distance calling patterns. Whether they make long distance calls or not, all use telecommunications infrastructure serving their premises. It's the same with the Internet as it replaces the publicly switched telephone network (PSTN) for voice calls and even cable TV for video. "Broadband usage" is no longer a meaningful metric.

If the calendar read 1999, the NTIA's report would be timely rather than more than decade out of date. Back then, "broadband" and "high speed Internet" was an emerging service option offered by legacy telephone and cable companies. Customers paid about $50 a month for the service over and above their usual monthly service charge.

Accordingly, discussing adoption of this service in terms of demographics and income would have made sense then since some groups of people would find this premium service more appealing and affordable than others -- especially since Internet applications such as websites and email were at the time only just starting to reach most consumers.

However, at a time when the Internet provides multiple services that formerly required separate, proprietary cable and telephone systems to deliver and can do so over a single tiny fiber optic strand connected to every home and business, reports like the one being issued today by the NTIA are increasingly irrelevant. It would be more far more useful and relevant if the NTIA and others instead studied how to hasten the build out of fiber optic infrastructure so that no homes and businesses are left offline.

Blair Levin perpetuates false distinction among IP-based services

Blair Levin, in another recent interview looking back on the U.S. National Broadband Plan he lead authored for the Federal Communications Commission before becoming an Aspen Institute fellow this summer, perpetuates a false distinction among Internet Protocol (IP)-based telecommunications services. IP-based services include Internet applications such as web browsing, email and e-commerce as well as Voice Over Internet Protocol (VOIP) and video, also known as Internet Protocol TV (IPTV).

In an interview with Marguerite Reardon of cnet news, Levin does so by differentiating VOIP and IPTV from Internet applications. Levin -- as do many incumbent legacy phone and cable companies -- continues to describe the latter as "broadband." That term was appropriate in the mid-1990s when "broadband" denoted a premium service offered by telephone companies over their single purpose, proprietary copper cable plants. But as fiber optic cable technology increasingly obsoletes metal wire for delivering multiple IP-based services, the term is no longer relevant.

Levin reinforces this artificial split by talking about "broadband adoption." That too was relevant in the 1990s when broadband was being offered as a premium service, requiring customers to sign up for or "adopt" it. Today, it no longer is when Internet applications, voice and video can be delivered to consumers over a single fiber "pipe."

Further reinforcing the bogus notion of "broadband adoption," Levin elaborates that "broadband" requires consumers to be literate whereas voice and video do not. Therefore, Levin implies, we first need to improve the literacy of Americans to drive "broadband adoption" before the nation revamps its outmoded telecom infrastructure with fiber. Here's what he told Reardon:

Even though there are a lot of low-income people who may not be able to afford multi-channel video (cable TV), there is still a high proportion of people subscribing to the service. And people are not leaving in huge numbers. The big difference between TV and broadband is that to watch TV, you don't have to be literate. The same is true of phone service. You don't need to be literate to use a cell phone, so penetration of those services is higher. But to use broadband for things, such as getting access to public services, health care, job training, etc., a basic level of literacy is necessary. It requires a skill set. And teaching people those skills is a serious effort. So price is a piece of it, but literacy and relevance are also aspects too.

This is so much sophistry. Moreover, even if one accepts Levin's false dichotomy between Internet applications on one hand and voice and video on the other, it would argue for a bigger push to deploy fiber optic telecom infrastructure since video requires the "fat pipe" bandwidth fiber can provide.

National Broadband Plan overly reliant on wireline, author says

Blair Levin, the Aspen Institute fellow who served as lead author of the U.S. Federal Communications Commission's National Broadband Plan before leaving the FCC this summer, told PCWorld last week the plan is flawed because it places too much emphasis on making landline Internet protocol-based telecommunications service accessible to all Americans.

"One of the problems we were running up against and that we should've been clearer about is that the conventional wisdom says the primary metric for measuring the validity or power of a national broadband plan is the speed of the wireline network to the most rural of residents," Levin is quoted as saying. "That way of looking at the problem is entirely wrong, is profoundly wrong -- almost every word in the sentence I just uttered is wrong. And we should've done a better job of explaining that."

If Levin could go back and rewrite the plan, landline and wireless technology would be framed synergistically, working in conjunction with each other to make a more complete telecommunications infrastructure that meets the National Broadband Plan's objective of expanding service availability to all Americans.

On this point, I agree with Levin. Until the last and middle miles of the U.S. telecommunications infrastructure can be fully upgraded to fiber, wireless has an important but interim role to play since it can be deployed more quickly than wireline plant. That's a very important consideration given that the FCC reported in late July that between 14 and 24 million Americans "still lack access to broadband, and the immediate prospects for deployment to them are bleak."

However, if Levin sees wireless connectivity as a replacement for fiber, I disagree. Wireless telecommunications is largely designed for mobile use and not to serve premises. Wireless also lacks fiber's ability to handle the exploding demand for bandwidth. There is no field-proven wireless technology that matches fiber's capacity to accommodate that growth.

As Tim Nulty, who believes fiber to the premises can pencil out even in rural areas, put it in a 2008 interview, fiber optic plant is to wireless as jumbo jets are to helicopters. "Think about 747s and helicopters,” Nulty told The Progressive magazine. “Helicopters are marvelous when they’re used for what they’re good at. But you don’t use them to fly thousands of people between Boston and Chicago. For that you need 747s.”

America's badly needed revamp of its telecommunications infrastructure should not be based on the expectation that wireless technology will overtake and render fiber wireline plant obsolete and cost ineffective. Hope is a good attitude, but does not a plan make.

Making fiber to premises a reality requires consumers to think like business owners

Much has been written on this blog and elsewhere about market failure and the urgent need for alternative business models to speed deployment of fiber to the premises telecom infrastructure. Most of it has been centered on market economics and technology.

However, a fundamental change in thinking must occur if these alternative business models are to come to fruition and bring the services people need now and in the future as bandwidth demand grows exponentially. People must think of themselves as not just consumers but also as owners.

Consumer cooperatives were formed in the U.S. a century ago to provide voice telephone service where investor owned telcos could not make a business case to provide service. Now that the telephone network is being replaced by the Internet, the time is at hand for the revival of this business model.

While coops offer significant structural cost savings that can make the business case pencil out for deploying an open access fiber to the premises network, those advantages cannot be realized until consumers think of themselves not just as a consumers but also as a business owners since a coop is a business, albeit owned by its customers. Being an owner requires doing diligence and assuming some degree of risk and not just asking what the coop may be able to provide them personally and at what price.

Without this shift in thinking, consumers will continue to be at the mercy of the incumbent telcos and cable companies and what services they choose to provide (or not provide as is often the case) and forced to pay whatever they want to charge for them in order to earn a return for their shareholders. Rather than benefit remote shareholders who could care less who gets fiber to the premises in their communities, it's time for consumers to say "enough" and take control of their telecommunications service.