|ILLUSTRATION: HOLLY LINDEM|
Utopia, as described by Sir Thomas More, the man who originated the term in the early 16th century, is an imaginary place of few laws, great natural abundance, and an absence of poverty and want. We still don't know how to cure poverty and want. But in a western U.S. desert, a utopia of sorts is taking shape for broadband users who would like to get their phone, television, and Internet services from the providers of their choice.
As it turns out, this Utopia, known formally as the Utah Telecommunication Open Infrastructure Agency, promises to be just that, a broadband utopia. And it is very much a real place, encompassing 14 cities in northeastern Utah. It delivers to each of its 3000 subscribers high-speed Internet access, telephony, and television programming through a fiber-optic cable at data rates that now reach 30 megabits per second. Soon, service providers there will be offering speeds of 50 and even 100 Mb/s. That's enough to download a 2-hour movie in about 6 minutes, 10 to 20 times as fast as the typical U.S. cable or digital subscriber line connection, 6 times as fast as Verizon Communications Inc.'s much-publicized fiber-to-the-home service (called FiOS) and twice as fast as the new DSL now being introduced in Europe by France Telecom and others.
The yellow-orange deserts, green cacti, and snowy white mountains of Utah's Wasatch Valley have been fertile ground for innovation since at least 1969, when a message-processing computer was installed on the campus of the University of Utah to serve as the fifth node on the Arpanet, the ancestor of today's Internet. Since then, the valley has spawned hundreds of high-tech companies, including Novell, WordPerfect, and the graphics firm Evans & Sutherland Computer.
The valley is dotted with some 30 cities, most of whose inhabitants are descendants of the Mormons who rode in on wagons in the 19th century. Today Interstate 15 connects the dots, and if you avoid rush hour and snowstorms, it's a 6-hour drive from Utopia's northernmost outpost at Tremonton to its southern tip, Cedar City [see map, "Broadband Corridor"]. Three-quarters of Utah's 2.5 million people live in the valley, but no single municipality has more than a small fraction. While Salt Lake City is the best known, its 185 000 residents constitute only about 10 percent of the valley's population, and it's not even twice the size of the second-largest municipality, its neighbor, West Valley City.
Utopia's promise is as vast as the nearby Great Salt Lake. In fact, if Utopia's data bits were liters of mountain water, the flow into one household connection would be enough to slake the thirst and bathe the bodies of half the people in the United States, whereas DSL could barely serve just Utah, and an old-fashioned dial-up connection would be enough for just a small town like Orem, population 84 000, the first of the 14 cities to be connected to Utopia.
What do you do with all this bandwidth? First off, you do just what you were doing before—only faster, better, and cheaper. Cheaper, of course, gets everyone's attention, especially people like Orem housewife Wendy Seamons de Hoyos. She and her husband, Ben, switched to one of Utopia's economy packages because it cost just US $44 a month, compared with the $69 per month they had been paying for DSL from Qwest Communications International Inc., the Denver-based incumbent regional phone provider. The cheaper package caps the bandwidth at 15 Mb/s, but that's still 10 times as fast as what Qwest had provided. At the new speed, de Hoyos says, Web pages "jump on-screen instantly."
The need for even more bandwidth will be apparent when new, data-hungry applications come onstream in the next few years. Most television today is a low-resolution digital version of the analog service of yore, but it's poised to soon bulk up its bandwidth requirements sevenfold for high-definition technology. Telephone calls will soon be as high-fidelity as FM radio, in stereo; that service, too, will require a fatter data pipe. Videoconferencing, still often a herky-jerky pantomime, will finally be easy and beautiful, experts say, and this will consume many megabits per second. In fact, high-bandwidth media will be pervasive—in online shopping, distance learning, telemedicine—you name it.
And in practice, it won't be so very hard to use 100 Mb/s. Utopia customers will be able to order the full-bore service, getting it all in a single, supersonic data feed, but most of them will prefer to divide this into separate streams. To split out one high-definition TV channel for the parents, another for the older children, and yet another for the youngsters, for example, at about 20 Mb/s each, would take more than half the feed. Reserve another 10 or so for high-definition video telephony, and there's only 30 Mb/s left for regular Internet use. This provision of television, Internet, and telephony services is called the triple play of consumer telecommunications.
Businesses stand to benefit from Utopia the most. A 20-employee accounting firm in the town of Murray now pays Utopia $150 per month for a 30-Mb/s Internet connection, down from the $650 per month it had previously paid for a 1.5-Mb/s line. And the impending 100-Mb/s speed is just the start. Utopia officials say they designed a lot of slack into the network, so that it can move up easily to 1-gigabit-per-second service over the next decade; indeed, business customers will soon be able to request this level.
The peaceful coexistence of multiple service providers is another thing that distinguishes Utopia. Because Utopia sends TV programming as Internet packets, indistinguishable from e-mail, Web pages, and everything else, it puts a huge reservoir of bandwidth at the disposal of its providers. By contrast, Verizon's FiOS, a sort of DSL on steroids, reserves most of an optical fiber's capacity for television, which means that FiOS customers who go to the open Internet, instead of to Verizon, for television programs have to cut into their Internet broadband, which is capped at 15 Mb/s.
Verizon's FiOS TV Service is second to none, offering 20 high-definition channels. By the end of 2006, it is expected to give 6 million homes some much-needed competition to the local cable provider. But that competition—two strong providers—may be all that those communities ever see. The de Hoyos family, on the other hand, can pick their TV service from a potentially long list of companies that vie equally for their dollars. "The open service-provider network model finally gives us a fair playing field to compete on," says Jon Hansen, CEO of MStar.Net LLC, based in Murray, Utah, which offers the triple play to Utopia customers for less than $100 per month.
It's still early Days for Utopia. When I visited in late January, the company was wiring up the sixth of its charter cities; in early March only about 3000 households had signed up for at least one service. There were four broadband providers and two for telephony, but only one for television. More can be expected, though, because Utopia encourages competition by having a business model that distinguishes among three things: the network itself, access to it, and the services that travel across it.
In that model, the physical network—the central offices and servers, the gray or green cabinets on concrete pedestals visible on neighborhood street corners, and the fiber itself, some of it buried under the streets, some overhead on existing electrical or telephone poles—is owned by Utopia, which is a nonprofit government agency that is considered an extension of the municipalities that created it. [See photos, "In the Neighborhood."] A for-profit company, DynamicCity Inc., in Lindon, Utah, operates the network under contract. DynamicCity oversees the contractors digging the ditches for the fiber lines, sees to it that the network's data traffic flows smoothly from its operations center, sends out repair and installation trucks when needed, and works with service providers such as the one the de Hoyos family now gets their broadband from, which happens to be XMission LC, of Salt Lake City.
Utopia's executive director, Paul Morris, compares the model to that of an airport. "It would be absurd for each airline to build its own airport," he says. "But that's just what we've been doing for telecommunications. Qwest has its set of wires in the ground, and Comcast"—the dominant cable provider in the region—"has its own. We think it makes sense for a city or a region to build the airport, have someone operate it, and let as many airlines provide service as want to." Besides the obvious efficiencies of the airport model, there's one other advantage—funding. A municipal project can be paid for with bonds, and bonds, like a mortgage, can take 20 years to mature. Happily, that's less than the half-life of the optical cable. What's more, the interest on municipal bonds is usually lower than commercial capital financing, even for a blue-chip company like Verizon.
Utopia isn't the only municipal fiber project in the world. There are quite a few around the globe, a number of them in Scandinavia. There are even a handful in the United States, including a five-year-old network in Grant County, Wash. Five years is a lifetime in the dog years by which such cutting-edge technology is measured. Utopia, though, might at the moment be the world's largest project of its type—at least until the Dutch build a fiber network covering all of Amsterdam, as they plan to do in the next few years.
Nor, it must be noted, is Utopia the only fiber project in Utah. In December, Qwest began serving a community called Daybreak. Unlike the Utopia cities, though, Daybreak is a new housing development, or "green field," and the lines that will go in might just as well be glass as copper. Placing new wires into soil that has already been tilled—an overbuild, in industry parlance—is harder to justify economically, especially when offering service to every household, as Utopia does. That's a commitment that most incumbent carriers—including Verizon—are so far refusing to make as they move from copper to glass.
In networking, bigger is almost always better. As a concession to that reality, a Utah law allows municipalities to group together to form what it calls "interlocal cooperation entities"—nonprofit corporations that benefit each municipality equally. Pooled emergency services for much of the valley, including every major city except Salt Lake City, are one example.
Such arrangements also pool knowledge. The city attorneys for these towns cannot hope to master all the intricacies of every bill in the legislature, so each tends to specialize in a few areas and shares that knowledge with the others. Through the 1990s, Paul Morris was the attorney for West Valley City. The executive director of the Utah League of Cities and Towns, in Salt Lake City, asked him to make telecommunications his specialty, because Morris was known to "like gadgets" [see photo, "The Lawman"].
Neither Morris nor West Valley City was leading the charge that would become Utopia, though, back in late 1999. Rather, the project got its start when representatives of two nearby Utah cities, Provo and Murray, returned from an American Public Power Association conference that was exploring the use of optical-fiber networks for meter-reading and control systems, using the same rights-of-way as municipal electrical networks. City officials saw a chance to piggyback some services of their own, such as central control of traffic lights and emergency communications.
Officials in Murray got the idea of building a municipally owned broadband network that the local electric utility could use. Provo, on the other hand, wanted a network to connect its 13 electrical substations and to extend cable TV service to the entire city, not just to the prosperous neighborhoods that the incumbent provider, TCI, was serving. Provo had franchised a second cable provider, Provo Cable, but it was nearing bankruptcy. Meanwhile, the town of Spanish Fork, near Provo, had already sold bonds to finance a municipal cable company. Earlier in 1999, TCI had been bought by AT&T, a much larger and more aggressive company. Then, in 2000, the city of Provo bought Provo Cable. AT&T, disturbed at the prospect of a city-owned competitor and eyeing developments in Spanish Fork as well, lobbied the state legislature for a law that would keep local governments from competing with the private sector. Morris, as the specialist in telecommunications, negotiated on behalf of the cities.
As a lawyer, Morris was slow to grasp the technical advantages of an optical-fiber network. But he quickly became convinced that fiber was the future for the area's high-tech businesses, for general business development, and for the average Utahan as well. "But it was hard to get the lawmakers to see the benefits of broadband, how it could generate economic activity or improve the quality of life," he says. "I remember a radiologist from Provo coming up and speaking about how a fast enough network could transmit an MRI to a specialist in Minnesota. That made an impression on some of them.
"At the last minute, some technology consultants working with the city of Murray, a company named Pinnacle, came forward with the open model, the airport model," Morris recalls. "So we got an exemption in the bill that says, basically, you can build a wholesale network, that is, you can build a municipal network, so long as you don't sell retail services. The city can't be the party selling cable television, or broadband, or phone service to a homeowner."
With that one provision, the new law, Utah's Municipal Cable Television and Public Telecommunications Services Act of 2001, went from being a firewall protecting incumbent providers from competition to being the enabling legislation for Utopia and an open door to far broader competition. Soon, 18 cities began talking among themselves about creating an interlocal cooperation entity for a single large-scale fiber network [see table, "The Original 18"].
Provo, preferring not to wait, was not one of them, and began converting Provo Cable's coaxial cable to an optical-fiber network that would come to be called iProvo. It retained the company's cable headend—the data center that receives TV feeds from the major studios and networks—HBO, CBS, NBC, ESPN, Discovery, and so on—but, in light of the new law, it could function only as a wholesale provider to retailers.
During 2003 and 2004, 11 city councils passed resolutions that guaranteed the bonds that would be issued by Utopia. Three others, Cedar City, Cedar Hills, and Riverton, remained in Utopia, but declined to guarantee the bonds. In April 2004, Salt Lake City decided not to lend its support, in a 5-to-2 council vote that followed heavy lobbying by Qwest. Three other founding cities also withdrew from Utopia during this period [see timeline, "And Then There Were Eleven"].
Ironically, one of Utopia's biggest supporters was AT&T, which by then had sold off its cable operations to Philadelphia-based Comcast Corp. AT&T had intended to sell the triple play, including Internet-based television, but so far the company offers its Utopia customers broadband service only.If not enough households sign up, 11 Utah cities will have to honor Utopia's bond commitments
In the summer of 2004, $85 million in bonds were sold to fund the first phase of construction: laying down fiber for the six southernmost cities. Utopia pays back its bonds by collecting a fee whenever a service provider signs up a homeowner for one of its services. The agency is intended to be self-sustaining, which it believes will be possible if 30 percent of the eligible homes subscribe to at least one service. If fewer households sign up, the 11 cities are on the hook, and they will have to honor Utopia's bond commitments with money from sales taxes that the referendums authorize them to collect.
Phase 2, in the five northern Utopia cities, began this spring and should be finished by the end of the year. Provo, the city that didn't want to wait for Utopia, wasn't expected to be fully wired until a month or so after Utopia's phase 1 cities—that is, in April. In a further irony, in 2005, iProvo's triple-play fiber provider, the first in the state, failed. Its customers are now served by MStar.
At first, Morris worked at Utopia for two or three days a week; later he took on the job full-time, taking a leave of absence from West Valley City. For a while, the city paid his salary; when Utopia became fully funded, it reimbursed West Valley City. He is one of only five full-time employees. The day-to-day running of the network is handled by what is, in effect, the prime contractor, DynamicCity, which grew out of Pinnacle, the consultancy that advised the city of Murray back in 2000.
Utopia differs from all other DSL and cable networks not just in its institutional aspects but in the technical ones as well. In its network design, the bandwidth going the last few hundred meters to an individual homeowner belongs to that household alone, whereas in a traditional broadband network, it is shared in the manner of a local area network.
The data travel from the central office, which is down the street from DynamicCity's headquarters in Lindon, passes through hubs of varying sizes, and eventually comes to one of many shoulder-high olive-green or battleship-gray cabinets at the side of the street. Each cabinet holds up to four access distribution switches and can serve as many as 224 households. These switches make the Utopia system especially powerful; they are computers in their own right, with individual laser-based controllers that distinguish the traffic to and from each household, so as to give each subscriber a distinct and inviolate connection to the Internet. In other words, each subscriber has a dedicated, active connection.
Utopia's "active Internet" architecture contrasts with the FiOS system, which channels bandwidth through something called a splitter/combiner to individual households. Because these splitter/combiners contain no intelligent electronics, the entire FiOS network design is called a passive optical network, or PON, even though, as in all networks, the end points—at the central office and at the household—are intelligent and active.
In addition to its dedicated bandwidth connections, Utopia has one other advantage. It relies on the Ethernet standard to carry Internet Protocol data packets all the way from the central office to the individual subscriber, without changing the format. This tactic greatly simplifies the network. Active Ethernet is also the protocol used by iProvo, and one instance of its advantage is the ease with which service providers on the Utopia network are able to use the iProvo television headend, something that would be difficult and expensive to do if either were PON-based.
Of the major carriers in the United States, only Verizon is building a fiber-to-the-home network—FTTH in industry jargon. AT&T Inc. (the company now called AT&T was until last year known as SBC Communications Inc.; after merging with AT&T, SBC took its name) and BellSouth (which is now being purchased by AT&T) are building what are called FTTN networks, where the N stands for "node." FTTN networks carry data on glass fiber only so far—the last stretch, from the node to the home, travels over the same copper wire that phone companies have used for a century. But the engineers working at Utopia's access provider, DynamicCity, say the differences between Verizon's FiOS and FTTN networks are overshadowed by the way in which they are alike—they are both PONs.
Verizon's engineers are quick to point out the advantages of their FTTH PON design. The principal one is cost. "In a star topology, like Utopia, there's one set of lasers and optical receivers for every customer served," says Brian Whitton, Verizon's executive director of broadband access technologies. "We can take the electronics in the central office and the hub and amortize it across the 32 hub customers." He also notes that while his system nominally limits the customer to a data rate of about 20 Mb/s—622 Mb/s going to the hub, divided by 32—it's rare that all 32 customers are using every last one of their available bits per second. When they're not, a customer needing more can temporarily exceed the nominal maximum rate. Finally, Whitton says, later this year, Verizon will begin to upgrade its network to Gigabit PON, which will increase the data rate to the hub to 2.4 Gb/s. That would give each customer 75 Mb/s. "Utopia will say, x2018But we're using cheaper lasers,' which is true, but they still have a considerable up-front cost," Whitton says. "Much of our cost per customer isn't incurred until they actually order the FiOS service."
The engineers at DynamicCity point out that many of Utopia's costs, such as the final wiring from the street into the home, as well as the in-house equipment, are also not incurred until a customer orders the service. In addition, Utopia is less concerned with whether costs are up-front or down the road, because its construction is financed with those 20-year bonds. Verizon, on the other hand, has already heard rumblings of mutiny from its shareholders, some of whom don't like the large capital outlays needed to upgrade neighborhoods to FiOS.
The real gains will come later on. Jeff Fishburn, DynamicCity's chief technology officer, maintains that as homeowner data rates climb toward 1 Gb/s, Utopia's cost per subscriber will remain nearly flat, while a PON's cost will climb steeply. According to Fishburn, Utopia's network would hardly have to change at all for every subscriber to have up to 1 Gb/s today. "Of course, our connection to the Internet as a whole would have to be upgraded," he says with a grin.
Then there's the upstream data rate to consider. This is the "a" in aDSL—asymmetric DSL—and the difference is enormous. Low-speed aDSL is 1500 kilobits per second down and 384 kb/s up; for high-speed aDSL, the numbers are 8 and 1.5 Mb/s, respectively. Even with VDSL—the "V" stands for very high bit-rate—a relatively speedy 13- to 52-Mb/s downstream rate is paired with a mere 1.5 to 2.3 Mb/s going out to the Internet. FiOS has a smaller difference, but it's still a 4:1 ratio: the aggregate data rate for 32 users, in megabits per second, is 622 down and 155 up.
That seems okay to the network designers at phone and cable companies—after all, most users download a lot of Web pages, for example, but rarely send them out. But from Japan to Scandinavia to Utah, the experience of symmetric networks is showing that when users are allowed to send at high speed, they tend to do so. To some extent, this reflects the behavior of peer-to-peer programs like BitTorrent, which mark such households as "superusers" and make them minidistribution hubs on the peer-to-peer network. Still, network activists point out the benefits of a design that favors active participation by all network users—bloggers over passive Web page readers, for example. Fishburn says that the Utopia design makes it easy for every high school to have, in effect, its own TV station. "And not just the school," he says. "Why not every high school student?"
Where does Utopia go from here? In a network, bigger is just about always better. At the moment, DynamicCity is building the network as quickly as possible, "footprint by footprint," Fishburn says. Each footprint passes 500 to 1000 homes, and at least two footprints, and as many as four, are added each week. Marketing has been, literally, door-to-door, with college students ringing doorbells, talking up the service, and taking orders.
DynamicCity would also like to attract more service providers for the triple play, and for other services as well. In January, it was experimenting with a videoconferencing setup made by Sony Corp., Tokyo; besides Sony's own software, DynamicCity was testing an application custom-made for it by a San Jose, Calif., developer, VSee Lab. "The original version of the software was written for low-bandwidth aDSL-style networks, but it took a lot of processing power and wouldn't run on a three-year-old Dell," Fishburn says. "Rewritten for a high-speed network, now it runs on any old computer." There are, he says, plenty of applications to come. "None of the home-security firms in the area are using the network, for example, but we're sure they will."
The executives at DynamicCity are already looking beyond Utopia's 11-city phases 1 and 2, to when the nonpledging cities' networks can be built out and still other cities can join Utopia. DynamicCity's long-haul fiber, which travels alongside I-15, starts in Idaho and runs south through all of Utah and well into Arizona. Any city near it can easily be brought in, and Utah law even allows interlocal cooperation entities to cross state borders.
The future may even take DynamicCity far from the I-15 corridor. Nate Taylor, the company's product development director, told me he spends most of his time outside the state, and Jeff Fishburn seems to do so as well. While the company officially declines to say where else it might form a similar entity, when I visited in January, Fishburn let slip that he had just arrived from Minnesota. That's not surprising—Minnesota represents the northeastern edge of a vast territory, consisting of 14 U.S. states from the Central Plains to the Rocky Mountains and beyond—where Qwest, the financially weakest of the four (soon to be three) surviving descendants of the original AT&T monopoly, is the incumbent telephone carrier. Qwest currently plans to build optical-fiber networks of its own in only two small areas—a few parts of its home state of Colorado, and the Daybreak housing development in Utah—opening the door for DynamicCity to convince other far-seeing cities to reproduce the Utopia model that it has helped define and refine. It's a model that cities around the world may well want to copy.
To Probe Further
The book Fiber to the Home: The New Empowerment, by Paul Green (Wiley, 2005), offers a detailed guide to optical-fiber networks.
Riverstone Networks, which manufactures some of the optical equipment used in the Utopia network, has a white paper on passive and active optical networks. "The Promise of the Triple Play" is available at http://www.riverstonenet.com/solutions/triple_play_promise.shtml.