In Big Cities

Building a communications network isn’t easy for anyone, not for Google, not for Comcast, and definitely not for cities that have never done it before. When the city of Seattle entered a partnership with telecom start-up Gigabit Squared in October 2012, hopes were high. The city and its partners were to connect 12 neighborhoods with gigabit-speed Internet service. In the following months, officials hyped the network, and officials like former Seattle Chief Technology Officer Erin Devoto heavily advertised the fact that the city had a ton of dark fiber and that they were ready to share it with the people who treaded above it each day. But then, deadlines were pushed back, problems with the partnership started popping up in the news, and twelve neighborhoods became just two. By January 2014, the project had completely fallen apart, and Gigabit Squared’s legacy to the city was an unpaid invoice of $52,250. Looking back, it’s easy to see how the project fell apart.

In fact, the episode turned one would-be municipal-broadband supporter into a skeptic. Robert Kangas, leader of Upping Technology for Underserved Neighbors (UPTUN), a Seattle broadband consumer advocacy group, said that watching failed projects like Gigabit Squared in Seattle, UTOPIA in Utah, and the municipal network in Provo, Utah, has cast serious doubt on whether big cities have the tools, knowledge and skills to run their own networks. That certainly seemed to be the case with the Gigabit Squared fiasco, Kangas said.

One of the biggest problems, Kangas said, was that the person who was in charge of managing the city’s technology, former CTO Erin Devoto, didn’t herself have a technology background. Next, the company that was to build the network, Gigabit Squared, had zero experience in developing a project like the one they were proposing, Kangas continued. What’s more, Kangas said, the city was trying to sell something it hadn’t even accounted for – the dark fiber that the city was bragging would transform the city wasn’t even available to be used because the city didn’t know which segments belonged to whom.

The breaking point in the relationship between Seattle and Gigabit Squared came when the city asked Gigabit Squared to fund a $100,000 engineering study to identify the details of the city’s existing fiber infrastructure. “It’s like the city just talked to a bunch of marketing people and said ‘we have this resource available, we want you to try and sell it, and we’ll figure out the details later,’” Kangas said, and “later” did come – it was the day the project got canned.

And while all of this was going on, Seattle already had incumbent providers competing in the market, but a Seattle Department of Transportation (SDOT) regulation called Director’s Rule 2-2009 was making it very difficult for anyone but Comcast to expand their networks in the city. When pressed about the rule, SDOT has repeatedly stated that it was created to prevent companies from placing ugly equipment boxes on public property. The city cited complaints it had received about unsightly utility boxes cluttering the environment. A public records request submitted by Kangas revealed that the city had received fewer than five complaints in eight years about existing public utility boxes. “I don’t think they realize their own policies are helping Comcast get a strangle on the city,” he said.

The city’s new mayor, Ed Murray, said in a blog post that his office is looking to examine “all options” for broadband in the city, and that they are considering changing the director’s rule. Until that happens, Kangas suggested that the city conduct an engineering study to identify where its dark fiber is, figure out who owns it, and make the data publicly available to attract investors in the market. Building a municipal network in Seattle without experienced partners would be almost impossible, he said.

“I don’t think the voters of Seattle would ever approve such a big expenditure based on the fact that we struggle to pass bus service measures and educational stuff or fire district initiatives,” he said. “Trying to sell to the voters something that would be $450 million to $800 million just to build up a network – this isn’t even maintenance costs we’re talking about – is a very, very difficult battle.”

How Did You Do That?

Chattanooga’s utility provider, EPB, has had outstanding success operating a county-wide fiber network serving 63,000 homes and businesses. Their success stems from several things, but it began with one man’s submission to humility. Harold DePriest was the new president of EPB in the late nineties (and still is today), and he was walking to have a chat with the new mayor, Jon Kinsey.

“When we were talking, Jon looked at me and said, ‘Harold, just what does EPB do for Chattanooga?’ and it really, really made me angry,” DePriest recalled. “Because I thought it was self-evident. We provide low-cost, reliable electricity, and I left his office really angry, just steaming, and then when I sat down and began to think about it, I realized what Jon was asking was a valid question, because other than selling electricity, we didn’t do much. We didn’t play a role in economic development, we were difficult to get along with, we were a typical electric company. We were reactive and slow and ponderous.”

EPB’s path had been altered. In the following months, there were changes in executive management and a big change in the organization’s outlook and culture. If it wasn’t good for the city, EPB didn’t do it. “It was things as simple as moving our poles quickly when they needed to widen the road,” DePriest explained. “Or back in those days, there were a lot of urban renewal projects, rebuilding neighborhoods and we just started looking for ways, instead of being a hindrance, that we could be of a help. And we found out we kind of liked that.”

When EPB committed itself to building a fiber network to support its electrical grid in 2007, their new outlook allowed them to think bigger. Smart grid monitoring turned out to be a great cost savings for the utility, but it was serving the citizens of Chattanooga and the outlying region with gigabit Internet service that put them over the top financially.

EPB’s success is as traditional as having a sound business plan that includes multiple income streams, DePriest said, but they did break the mold in one way. The usual procedure for building a network is to start where the money is. Studies are conducted, estimations are made about what percentage of various neighborhoods are likely to become customers if service is made available, and then the network is deployed in the areas with the highest estimated take rates. Chattanooga’s fiber network was built in the poorest areas first.

“Frankly, what we found is that the take rate, if you’ve got a reasonable price, doesn’t vary a whole lot from the poorer areas to the richer areas,” he said. “The difference is that poorer people tend to live in smaller houses on smaller lots. Richer people live in great big houses on great big lots. There are a whole lot more people that you can get to quickly [in poorer areas] and we’ve not seen a whole lot of difference on the tendency to buy our product based on the income of the area that we’re serving. In fact, our sales are pretty uniform all the way across our system. I think the way to serve people is to have a product that’s reasonably priced that they actually want to buy and use. I think it is that simple.”

Some point to the $111.5 million in federal stimulus funding as an explanation of Chattanooga’s success, but that funding just shaved seven years off their build time, DePriest said – it didn’t change how they ran their business. And operating a fiber network is a business with a lot of office work, he said.

“Everybody thinks in terms of engineering technicians, the technology, and there is some validity to that – this is some pretty complex technology,” he said. “That was never my biggest concern. My biggest concern was all the back office stuff necessary to make a business work. How do you generate the bills? How do you talk with customers? How do you make a sale? How do you hook them up? How do you deal with them when they want to change things? Those back office things, it turns out, we do them on the electric side all the time.”

EPB is a successful fiber provider because they have been supplying a similar service, electricity, since 1935. “For example, we send out 175,000 bills every month on the electric side,” DePriest said. “So figuring out how to do the billing for another 60,000 fiber customers was difficult, but it wasn’t impossible. On the other hand, if we didn’t know anything about billing, boy, that would be a big job.”

When people talk about government staying out of telecom, DePriest said he thinks back to EPB’s origins in the 1930s. “Only three percent of Tennessee farmers had electricity,” he said. “They could have waited decades on private industry to get there, or they could create municipal systems like Chattanooga, EPB, or cooperatives. I’m pretty open to the notion if this infrastructure is important to a community, why doesn’t that community have the right to provide the infrastructure for themselves?

“Every year in the state legislature, we’re trying to expand and our competitors are trying to build walls around us. But, at the same time, they don’t seem to want to serve the people in some of these rural areas. And we as a country are going to have to figure out if we want people all around the country to have this type of service. I find it hard to understand why we’d keep anyone from doing it. If the goal is for Americans to have better Internet, better communications, then there's room for anybody who can work toward that goal.”

Editor's Note: Comcast and Time Warner Cable did not respond to interview requests for this story. 

Colin Wood Colin Wood  |  Staff Writer

Colin has been writing for Government Technology since 2010. He lives in Seattle with his wife and their dog. He can be reached at cwood@govtech.com and on Google+.