The LUS Fiber system, a municipally-owned fiber project started taking customers in early 2009. Over the last three months, the books have tipped to cash-positive. That is, revenue exceeds bond payments and operating expenses. That means they can either lower prices, or accelerate bond payments. It is a validation of the model of public ownership of communications infrastructure that incumbent carriers, such as the cable and phone franchisees, would rather you never, ever heard about. It is well past time for Portlanders to seriously consider how we get there as well. Ask your candidates where they stand on the issue. Point them at the model we prefer.
This video addresses what is at stake in our communications future here in Portland and around the world.
This morning, I sat through most of a presentation/discussion of the draft Portland Broadband Strategic Plan with a number of invited telco industry representatives at City Hall’s Council Chambers. The representatives were from CenturyLink (formerly Qwest), Comcast, EasyStreet, Integra, TW Telecom, LS Networks, and Don Westlight as a representative from NWAX, a local peering exchange. They all said predictable things. CenturyLink thought the public sector should stay out of the infrastructure business. They all touted their highest speed capabilities without mentioning any pricing or terms-of-service constraints. The competitive companies talked about their ability to bring metro ethernet to businesses. They all agreed that for business services, Portland was a competitive market. When asked directly about residential competition, they talked about how competitive the business market was.
Rich Bader of Easy Street made the obvious point that businesses care about their finances and making money, and that the public policy wants infinite capacity for free for its productivity benefits to society, and that everybody should recognize that tension. There was a discussion about access to open trenches for installing conduit. Someone suggested that when someone gets a permit for a trench, that notice should be given to other franchisees so that they have the chance to drop conduit into those trenches while they are still open. Don Westlight pointed out the free-rider effect might cause people to stand around waiting for someone else to dig the trench under those circumstances and that there should be a way of sharing the costs of digging the trench. There was a discussion about access to buildings, and how some property owners want fees from the carrier to serve tenants in their building. Someone said the way to get more development of infrastructure was to keep costs low, by avoiding redundant work, or by building in access when construction costs are relatively low.
I left with the feeling that these companies had touched on the problem, but had not generalized it broadly enough. They are all talking about figuring out how to serve a customer who lives in Beaverton and wants to get to the airport, and figuring out the cheapest way to build a highway in between for their customer to use. The obvious answer is, SHARE THE INFRASTRUCTURE! Build an infrastructure with massive capacity and share access to it. All of these companies are working around the broken model that they have inherited of dinosaur-like incumbents with a lock on the infrastructure. The smaller competitive companies are only filling the holes where the dinosaur can’t or won’t reach, where the need is so great that people with resources are willing to pay through the nose.
After listening to them talk, I think the case is even stronger for public ownership of the last mile infrastructure. It’s cheapest, because it means building one network of fiber, and not a dozen. And it means equitable access to the massive capacity that infrastructure provides.
Portland needs to invest in scarcity-squashing infrastructure. The City has built its own fiber infrastructure for governmental use. It’s time for the City to look out for its citizens and let them own their own infrastructure as well. The public stands to reap enormous benefits, both in avoiding perpetual rent and by enabling competition in service providers on the infrastructure. Don’t let this opportunity slip by.
There was a recent piece on a PBS program called “Need to Know”, very much worth a watch.
It shows the kind of competition spurred by Open Access rules on infrastructure. Sadly, the FCC is such a basket case, we are unlikely to ever get these kind of sensible rules if we have to rely on them. A recent article by Benoît Felten explains why. Localities are more apt to understand the urgency and aren’t as insulated from the failures of the existing model.
After about a year of deliberation, Google has announced their choice of Kansas City, Kansas as the test-bed for their 1Gbps fiber-to-the-home network. Congratulations to Kansas City and Google for moving forward with their project. We share their excitement about what the future holds there.
Meanwhile, the City of Portland has been considering what it can do for Portlanders. Since last autumn, the Office of Cable Communications and Franchise Management has been working on a Broadband Strategic Plan. In January, along with dozens of others, I was asked to participate on a Work Group to help shape the Plan. The third meeting of this group of participants concluded on Tuesday afternoon. In the next few months meetings will be held to explain the draft and to take public input. On September 14th, a finalized Plan will be presented to the Portland City Council for adoption.
As of Tuesday the draft Plan was, in my considered opinion, much much too timid—embarrassingly timid—and not at all likely to meet the stated goals for competition and performance in broadband service in the next decade. From the first meeting of the group, I advocated a goal of publicly-owned fiber to every residence and business in Portland. I say that not because I am a wild-eyed Socialist, but because it is the only path likely to give us the competition and capacity at a reasonable price that we all want, and I’ll explain why.
As network users, competition is our friend. It keeps the service provider honest and their profits reasonable while providing an attractive and useful product. Competition is great. Unfortunately, the kind of competition we have is broken and inadequate. Presently, we have competition mostly between types of service. Cable vs DSL vs Wireless vs Dialup. Within each type of service, our choices are limited or (in the case of Cable and the newer/faster DSL) non-existent. There are two reasons for this. First, wireline infrastructure is expensive to build. Once it exists, the economics work against another carrier trying to enter the same niche. Consequently, we have one cable company, we have one phone company and, in the suburbs, we have one fiber-to-the-home company. Now, ideally, all sorts of service providers would be able to sell their services over each kind of infrastructure, so that network users could choose which ones met their needs best. That is at least partly how old-style DSL works. If you can get service at all, you usually have a fairly robust set of choices of Internet service provider (ISP) on DSL (mostly for historical regulatory reasons). This is called Open-Access. However, newer infrastructure is not regulated that way. The City of Portland in the late 1990s tried to impose Open-Access rules on the cable company. The cable company responded by saying, “No, this is our infrastructure, we own it, and we’ll do with it what we like,” and went to court and ultimately succeeded in overturning the Open-Access rule that the City had tried to impose. As a result, there is no competition on the cable system. If you want the bandwidth you can get with cable, you buy it from Comcast or nobody. If you don’t like their terms of service, you are out of luck. Take it or leave it. That is not competition. If you want competition, you need Open-Access.
So, how do you get Open-Access? Well, you could have the Federal Communications Commission or the US Congress mandate it. Unfortunately, neither of those institutions, ostensibly concerned with the Public Interest, shows much willingness or inclination to actually act in the Public Interest, and anyway, a Federal Court might ultimately decide against the Public Interest, as they sometimes do. Breaking up vertically-integrated telecommunications companies into infrastructure and content/service components could also provide the needed competition. However, that doesn’t seem particularly likely either. We have monopolies and no one seems to want to regulate them. So, what are we left with? If you are reliant on owners of infrastructure to do what you want, and they are disinclined to do it when they are someone else, you need to become the owner. Then, as owner, you can decide to do what you want, which is to provide Open-Access and Competition on the infrastructure. So, in short, Competition requires Open-Access, and Open-Access requires Public Ownership.*
How do we get Public Ownership? Well, we can either buy it from someone who has it, or we can build new. Because fiber-optic networks have massive capacity, sufficient to support today’s and tomorrow’s bandwidth needs on into the distant future, are already demonstrably practical and actively being deployed around the country and around the world, it makes most sense to invest the money on building a fiber-to-the-premises network with local public ownership and accountability to its users.
What would such a network look like? Here are the key principles:
- That it be operated in the interests of the users and that users have a role in governance;
- That users pay the costs of construction and operation and reap the full benefits of ownership;
- That users can connect to whomever is also connected to the network, on a consensual basis;
- That 1Gbps is available to any other end-point on the network;
- That any service providers can have access to the network at non-discriminatory rates;
- That services can be purchased from anyone else on the network, including but not limited to Internet connectivity, video or voice service;
- That privacy of traffic crossing the network is ensured, and that any surveillance is authorized by court-ordered warrants;
- Network operations centers are open to public inspection.
Isn’t public ownership un-American? Well, not really. There are already 133 municipal networks in the United States.
What about costs? Isn’t fiber expensive to build? Actually, it is cheaper than copper infrastructure. A City of Portland feasibility study a few years ago estimated it would cost about $450 million to build a fiber-to-the-home network throughout the city. That seems like a lot of money, however today Portlanders are spending $100s of millions annually on telecommunications services. The money is out there, it just happens to be going into a rathole. We are talking about spending 10 times that on a bridge over the Columbia!
Still, it sounds scary, and there is the possibility of making costly errors if we aren’t careful. I suggest that we build a demonstration project first, with perhaps 2000 homes and businesses somewhere in the city, which could be built for a few million dollars. We could even spice things up by having a Google-like selection process, to get people excited about it. When the demonstration project has shown its viability, we can continue growing the network until the entire city is served. Portland has a proud history of large capital construction projects that serve the public interest over the long-term. In the 1800′s, Portland had private water utilities peddling an inferior product. Together we wisely decided to make a better choice. Our Bull Run water supply was very expensive to build. We sold bonds, and through user fees we paid those bonds back. As a result 100+ years later, we have excellent quality water and a reasonable price.
We can be wise again.
Let the City know you expect the problems of Competition and Capacity to be addressed aggressively in their Broadband Strategic Plan. Don’t let us settle for more of the status quo.
* The “Open-Access requires Public Ownership” relationship does not hold true for the Google Fiber project, which is what made it so attractive. Google has explicitly stated that the network they build will be Open-Access.
* (as of June 4, some dates have been edited to reflect current schedules)
Forget 1Gbit, USA Today reports that a Swedish woman has an experimental 40Gbps fiber-optic Internet connection. While the article makes it clear that she only uses a fraction of the speed available to her, it’s important because it demonstrates the power of a fiber optic infrastructure.
The 40Gbps speeds were accomplished with a new modulation technique—meaning the light pulses traveling over the fiber are upgraded, but the fiber optic-cable itself does not need to be replaced. Fiber-optic cables carry light, which travels at the highest speed possible, period. Bandwidth increases are accomplished by altering the frequency and pulses of the light sent over the cable. These upgrades can be made without replacing the entire infrastructure. Unless and until someone invents ansible-like communcations, fiber-optic infrastructure will remain the fastest possible means of transmitting data.
Today, a friend of mine spotted Comcast in his neighborhood running fiber-optic cable down the street, but not replacing the copper infrastructure that runs to homes. Halfway build-outs like this will continue to necessitate replacement and upgrades to infrastructure. Incumbent providers expand their networks on the cheap, while keeping customers in the slow-lane of the Information Super Highway.
The United States Court of Appeals ruled today that the FCC does not have authority to require Net Neutrality on privately owned networks like Comcast’s cable Internet or Verizon’s FiOS. A variety of approaches suggest themselves to preserve the essential freedom of the Internet. One is to change the law to provide the FCC the needed authority, however this is likely to lead to protracted legislative and legal battles, where the network owners fight to preserve their control and freedom to manipulate conditions of use. Considering it is their investment, the private owners have some reasonable grounds for that position. Another approach is for the public interest to exert itself through direct investment. That is, for the public to build their own last-mile infrastructure and for that public infrastructure to be operated in their own interests, including freedom to use that infrastructure as they see fit, consistent with the physical limitations and fairness. Fiber optics to the end-user provides the best capacity bang for the buck. Public ownership provides the best freedom bang for the buck.
Someone on twitter recently mentioned Molalla Communications Cooperative as a nearby existing fiber-to-the-premises network. It is, and there are others. It has better bandwidth and pricing than we get from incumbents here in Portland, their X-Fon product has 40 Mbps down, 10 Mbps up for under $50/month. However, looking at their fine print, we find this clause:
“You shall not connect servers of any type to the Services. Molalla Communications reserves the right to suspend or terminate Service without advance warning if a violation of this policy is detected.”
If you take that literally, and I think you have to, it would mean no Linux boxes that you can log into remotely, no web servers, no business email servers, no other “normal” internet services. It prevents you from using off-site system administrators to help you solve technical problems. In many realistic circumstances, this restriction makes the service unacceptably limited. Comcast has similarly restrictive terms of service. Service providers seem to like the bright line between producers and consumers, that you are a consumer and a consumer only. “We went to the trouble of labelling you, stay in your damned box!”
Users of infrastructure should pay for the infrastructure, but having done so, they should have the ability and freedom to use that infrastructure for what they want, not limited to some artificial set of consumption-only activities. Ward Cunningham tells the story of how he changed the world because he could provide a service over a 14,400 baud modem with a $300 computer under his desk that let people communally develop and refine ideas about computer programming. The “no servers” rule would strangle in the cradle just that kind of innovation and collaboration that can make progress possible and life for everyone better. Maybe there are no creative people in Molalla (I know that’s not true), but there are in Portland. A Portland fiber network cannot have a restriction like that and be considered a success. This is why an open-access model, where many providers on the same infrastructure compete on price, quality and terms-of-service is essential.
Qwest, our local phone company–the company that owns all of the phone lines in the city, ran this full page ad in the Oregonian. Qwest is eager to advertise their “Fiber-optic* fast connection speeds up to 20 Mbps.”
What Qwest is less eager to advertise is the little asterisk on “fiber-optic” which leads to the fine print “Fiber-optics exist only from the neighborhood terminal to the Internet.” So, if you happen to live in the neighborhood node, this is a pretty great deal. Unfortunately, the node in my neighborhood is about 4 or 5 blocks away, and the size of a doghouse. Really, Qwest is selling plain old DSL, but they’ve moved the DSLAM (the device that creates the DSL signal on the phone line) into a box in your neighborhood, and connected it back to their office with fiber-optics. They could just as easily call their service “Ethernet fast connections,” because somewhere, far away from your home, they use Ethernet to connect equipment to the Internet.
Don’t be fooled by DSL service disguised as fiber-optics. The only truly fiber-optic service is one where the connection between your home or business all the way back to the data-center is fiber-optic. That is the only fiber service that will offer 1Gbps speeds. The best fiber service will also be open-access, allowing any service provider to compete to offer you the services you want. Qwest’s fake fiber doesn’t provide any of that, but open-access, community fiber would.
Qwest’s ad calls their service “100% fast. 50% off.” It would be more accurate to call it 50% fast, and 100% off. Qwest doesn’t offer fiber speeds or fiber service and they never will. That’s one reason Portland responded to Google’s Fiber for Communities RFI, and it’s one reason that we shouldn’t be content until we have a citywide, open-access, fiber to the home infrastructure.