Gig.U Asks Universites and Telcos To Work Together for the Internet of the Future
BY Nick Judd | Thursday, September 15 2011
A new initiative seeks to create "testbeds" for extraordinarily high-speed Internet access in partnership with 37 universities across the U.S., an attempt to succeed where federal policymaking efforts have so far failed to generate innovative approaches to Internet infrastructure.
That initiative, called Gig.U, was expected Thursday to release its first public call for information from vendors who want to start talking about how to make this project work. Championed by the chief architect of the National Broadband Plan, a 376-page policy framework released last year, Gig.U asks vendors to come up with ways to develop experimental networks at speeds of one gigabit, or 1,000 megabits, per second, in partnership with community coalitions anchored by universities. The initiative would provide Internet access 100 times as fast as the 10Mbps speed many providers offer as part of current middle-tier access common in American homes.
It also happens to not only step squarely into a large, expensive, and heated ideological debate over the future ownership of the Internet — it also asks some of the main combatants to come to the same table and cooperate. But more on that later. First, what "gigabit Internet" means and why it's worth reading about.
At gigabit speed, I could download a new high-definition copy of the 2009 movie "Star Trek," a nearly four-gigabyte file, in far less than one minute — but that's not the kind of use Gig.U is hoping to test out. A person who requires complex medical equipment to monitor a health condition could also keep those instruments in her home, and transmit readings securely to a hospital. A small business could have a connection to its off-site cloud computing resources, or a data center, that was so fast nobody would know the servers weren't in the building. A community center could handle a video conference with several incoming and outgoing video feeds. Applications for health and for video conferencing have already been seen in a testbed community operated by researchers at Cleveland's Case Western Reserve University called the Case Connection Zone.
But the truth is that nobody really knows how much people could do with ubiquitous Internet access at those speeds because right now so few people have it. Even in a place like South Korea, where fiber-optic home connections to the Internet are ubiquitous and cost less than Americans pay for their current plans, a nationwide gigabit fiber-to-the-home initiative began as a pilot only earlier this year.
"As a policymaker you're asked to invest in something that you can't see," explained Rudolf van der Berg, a policy analyst with the Organization for Economic Cooperation and Development based in Paris, "but you know in your gut, at the micro and the meso-level, has an impact on the way that your country works."
Governments like the one in South Korea and in some Scandinavian countries are making that bet without having concrete data on what the impacts will be, the way the Pentagon and American universities bet on building the redundant packet exchange network that evolved into the Internet as we know it today. The risk is that the next innovations will happen there, not here. Companies seeking a competitive edge online might go there, not here. And citizens will benefit from greater efficiencies and new ways of working there, but not here, because absent any real incentive for telecommunications firms to spend the money to build such a high-speed service, some experts argue, and without government intervention, there's no one to develop that kind of ubiquitous next-generation service.
For a country staking its future on innovation while lagging behind many other developed countries in terms of broadband access and speed, that's a raw deal. And it's the problem Gig.U is trying to solve, by building demand for super-high-speed Internet access and providing a glimpse into how it might be put to use.
Gig.U's champion is Blair Levin, a former Federal Communications Commission policy strategist who led the creation of the FCC's National Broadband Plan and served from 1993 to 1997 as the FCC's chief of staff under Chairman Reed Hundt. Levin, who cautioned that the plan was launched "in beta" and has publicly suggested alternatives to several suggestions he made in its first release, moved to the Aspen Institute after leaving the FCC; from there, he has been working with Case Western Reserve University's chief information officer, Lev Gonick, to build a coalition of universities and towns to open their doors for the Gig.U experiment. An attorney with a penchant for jokes about old TV shows, Levin says Gig.U is a "spinoff" of the National Broadband Plan. ("Like Frasier to Cheers," he says.)
"You need an ecosystem that's constantly improving," Levin told me in a phone interview earlier this week. "In order to have that, you need to have somebody at the high end doing leading-edge stuff."
In the National Broadband Plan, Levin called on the government to use military bases as epicenters for that kind of work. That didn't pan out, so Levin is now pushing instead for college towns, with their business incubators, engineering students, researchers — and, in many cases, existing fiber networks.
That's in part because most places in the United States are nowhere close to the leading edge. Per capita, fewer of us have broadband than in South Korea, Japan, or just about anywhere in Europe. Your service provider probably says you're paying for "up to" a certain speed — maybe 5 Mbps, or, if you play a lot of World of Warcraft, maybe 10 — but serves you far less most of the time. (That's what providers mean when they say your connection is "burstable" to a given speed.) You'll get a fraction of that if you're uploading data rather than downloading. If you don't like it, you can take your business ... nowhere, as many communities have at most two high-speed Internet service providers to choose from, and their prices and service offerings don’t vary much. And while this isn't and will never be France, consumers in some parts of that country can get Internet access ten times as fast as what I get in my home in Brooklyn, and for about the same price.
Adoption rates are changing. In keeping with the play, recommended in the Broadband Plan, to increase adoption through wireless Internet access and to grow the availability of broadband, the FCC is encouraging firms to bring broadband to rural and low-income areas. A commitment from Comcast to bring affordable Internet access to low-income areas around the country was a requirement in the FCC approval of the Comcast-NBC Universal merger, for example. Comcast is making good on that with a program called Internet Essentials. And the American Internet industry is getting faster — a new technology called DOCSIS 3 can deliver speeds of 105Mbps — but that's still just 10 percent of the speeds Gig.U is talking about, and an Engadget report from April pegs the expected cost at around $105 per month.
The recommendations in the National Broadband Plan didn't do enough to accelerate the development of high-speed Internet, critics say.
"As it turns out, the plan, what the plan recommended is basically what the United States is going to have anyway, at least for terrestrial wired networks for sure," David Isenberg, the former Bell Labs researcher and author of "The Rise of the Stupid Network," explained to me recently. "It was extremely un-visionary."
One of the forces that keep the current state of the American Internet so far behind the times, critics like Isenberg say, is the inertia of big telecommunications companies.
Those firms are already making plenty of money being big telecommunications companies, van der Berg, the OECD analyst, explained to me. Absent a disruption to the status quo, he says, an investment in an unproven technology makes no business sense for them.
"If you're a CFO of a large telco," asks van der Berg, "Why would you authorize any large scale investment? It is not yielding you money in five years at the level you're used to with your current network."
Inspired in part by Google's fiber initiative, which will bring fiber-optic Internet infrastructure communities across the country beginning with Kansas City, Gig.U might be the disruption needed to alter that equation. Each of its 37 member institutions will work with vendors to give gigabit Internet service a shot in their college towns. That might mean universities helping to clear red tape and stimulating demand, as the Los Angeles Times suggested when the initiative was first announced in July. It could also mean making arrangements to tap the unused high-speed assets of universities, hospitals, libraries and schools, creating a core infrastructure that a vendor could more cheaply and easily build around instead of installing all new cables from scratch. One prototype is the Case Connection Zone, the experimental gigabit initiative around Case Western Reserve in Cleveland. It could even mean a municipally owned service provider, especially in cities where there's a history of city-owned municipal utilities. Gonick calls this a "mayor-proof" structure that gives Gig.U flexibility and, hopefully, longevity.
The variety of people expected to start talking at the same table is surprising because nonprofit providers, big telecommunications firms and universities have until now been frequent political adversaries when it comes to Internet infrastructure.
Chris Mitchell is an advocate for municipally owned broadband networks, and consults with municipalities who are looking to start their own government-operated networks.
"The big companies like AT&T and Comcast absolutely don't want to be regulated and they don't want to deal with competition from the public," Mitchell, of the Institute for Local Self-Reliance, told me recently. "They don't want to deal with any competition, to be honest ... because it will lower their profits."
These companies are big spenders on Capitol Hill. AT&T spent over $15 million lobbying the federal government last year, according to OpenSecrets.org. Comcast spent nearly $13 million. And while those numbers are impressive, the real action is in the states: In North Carolina this year, advocates singled out Time Warner Cable in particular for pushing for a law that makes it more difficult for municipalities to form their own networks. In Wisconsin, a statewide university network battled this year with legislative attempts to cut millions of dollars in funding — an effort backed by the state's telecommunications lobby. Stories like this are persistent through time. Tim Wu, the technology journalist, law professor, and current chairman of the free-and-open-Internet-advocating organization Free Press, provides hundreds of pages' worth in his book Master Switch, released last year.
Hence the tension: When it comes to providing Internet service, universities and municipal broadband advocates have been in the same room with telecommunications companies most often when the room also has a judge and lawyers. People like Mitchell argue that telcos will do whatever it takes to maintain exclusive control of the physical cables and hardware — literally, the tubes — that make the Internet possible. Telcos argue in turn that universities and municipalities can make Internet service available at a lower price than they can, which they say stifles their ability to compete. (Never mind it is the telcos who have market dominance — but that's an argument for another day.)
"Often a provider's first reaction to something that has the words 'community,' 'university,' and 'broadband' together is not positive," Elise Kohn, the Gig.U project manager and a former FCC staffer, told me during a phone interview earlier this summer.
"There's a lot of history, and the history is not irrelevant," Levin admitted in my interview with him, "but we want to put it aside and say that what we want everybody to come with to this initiative is a sense of innovation."
With that in mind, the initiative has approached just about every major telecommunications firm in the country about working with universities to create these local, gigabit-speed networks, Levin said. They've responded with interest, but it won't be until Gig.U receives responses to Thursday's request for information that it will be entirely clear if that was genuine enthusiasm or politesse.
Jim Baller, a lawyer who works with municipalities on fiber build-outs, says the structure of the Gig.U initiative leaves the door open for telcos to retain ownership of the networks at the end of the day — meaning, in short, this isn't necessarily competition for them. He says that's not a bad thing because it at least brings them to the table — remember, major telcos likely wouldn't be talking about retail-scale gigabit Internet otherwise.
"I believe that GigU should not be viewed as either public-versus-private or as .edu versus telecom," he wrote to me in an email. "Rather, it seeks to simulate robust discussion of new ideas and business models from all directions. While GigU has explicitly encouraged established carriers to get involved, it is open to providers of all kinds."
Later in the email, he continued:
Is there a risk that the carriers will hijack the GigU project? Sure. The best way for other interested parties to avoid this is to participate actively in the process. With such participation, I believe that the risks are well worth taking ... At worst, if the project fails, we can take what we have learned and go our separate ways – which may include developing more community broadband projects.
In other words, Gig.U is an opportunity to get people talking about differing ideas about how the Internet of the future should look, who should own it, who should pay for it and how much it should cost.
And it begins Thursday in earnest with a request for information Gig.U is expected to release to gather comments, conceptual frameworks on how new gigabit Internet networks might look, and indications of interest from potential vendors. There will be an information session on the initiative on Sept. 26 at the University of Chicago, one of the partner universities in the initiative.
This post has been updated; on Thursday Gig.U announced that the roster of involved universities grew from 35 to 37.