The digital home


Published on

  • Be the first to comment

  • Be the first to like this

No Downloads
Total views
On SlideShare
From Embeds
Number of Embeds
Embeds 0
No embeds

No notes for slide

The digital home

  1. 1. The digital home Science fiction? Sep 1st 2005 | SAN FRANCISCO From The Economist print edition Technology firms are pushing a futuristic vision of home entertainment not because consumers are desperate for it but because they themselves are RECENTLY, at one of the fast-proliferating conferences devoted to the “digital home”, John Burke, an executive at Motorola, a maker of mobile phones and digital gadgets, showed a video that presented his company's version of this vision. In the clip, a youngish man wakes up to a rock video that automatically starts playing on a screen next to his bed. He gets up to have breakfast and the rock video follows him to a screen in the kitchen. He moves into the living room and up pops the rock video on yet another screen. When he leaves his flat and gets into his car, the video starts playing on a screen in the steering wheel.
  2. 2. To ordinary humans this sort of thing must seem like silly—or downright frightening—marketing claptrap. In fact, even Mr Burke's audience of self- selected technophiles seemed sceptical. “Did you notice that the guy was a bachelor,” said Tim Dowling, the boss of Pure Networks, a software firm in Seattle that helps users to set up and troubleshoot home-computer networks. “That alone tells you that they're out of touch. I thought: How dumb.” Real people do not want to be hounded through their home and their life by some video stream, he argues; they just want help with basic headaches, such as getting the kids' laptop, mom's Apple Macintosh and dad's Windows machine to share the family's printer. Whether or not computer, software, consumer-electronics, telecoms, cable and internet companies are in fact out of touch with consumers may be the biggest question facing these industries today. That is because the “digital home”, a concept and category hugely hyped in executive circles but still rarely heard in discussions among consumers, represents their greatest hope for revenue growth. Demand from corporate buyers of technology has barely recovered from the dotcom bust and is widely expected to be unimpressive for years. By contrast, the homes of consumers appear to technology vendors as a barely tamed analogue wilderness. Darcy Travlos, an analyst at CreditSights, a research firm, estimates the market opportunity of the digital home at $250 billion in America alone and $1 trillion worldwide in three to seven years. “We view the digital home as critically important,” says Craig Mundie, one of three chief technology officers at Microsoft, the world's largest software company. “The home is much more exciting than the workplace.” Computers have already led to small revolutions in boosting productivity in the office and helping people to communicate and to be creative, he says, so “we're pretty confident” that computers will have a similar effect on the way people consume entertainment. Intel, the world's largest semiconductor maker, recently reorganised itself into new business divisions including, prominently, one called “digital home”. Last week it formally launched Viiv, a bundle of chips intended for use in digital-home PCs. Consumer-electronics firms such as Sony, computer-makers such as Hewlett-Packard (HP) and Apple, telecoms giants such as Verizon or SBC, cable companies such as Comcast, internet firms such as Yahoo!, networking- equipment companies such as Cisco—all agree that the digital home is where the action will be and are investing furiously to make sure they have a good chance of playing a leading role.
  3. 3. Their first challenge in stimulating any sort of consumer interest is the difficulty of merely explaining what the digital home is supposed to be. You might think, for instance, that the term refers to the long-established trend away from analogue and towards digital media. In music, most people have completed their migration from vinyl records and tapes to digital CDs. In films, the trend from videotapes to DVDs is not far behind. In photography, traditional film is fast being replaced by digital cameras and pictures. TV and radio broadcasters are also shifting to digital transmissions, with Britain leading the way. Confusingly, however, that is not what vendors mean when they talk about the digital home. Instead, they invariably mean a home in which all sorts of electronic devices—from the personal computer (PC) to the TV set-top box, the stereo, the game console and, in some versions, even the garage door and refrigerator—are connected, both to one another and to the internet. Hence the Motorola marketing video that Mr Burke was showing. Its purpose was to illustrate what Motorola, like Microsoft, calls “seamlessness”, as digital content hops automatically between various devices and screens. The excitement, therefore, is not so much about content being digital, but about its delivery switching from physical things (such as CDs) to photons (such as wireless downloads or streaming), because this requires consumers to buy new gadgets. Believers in this future point to encouraging statistics. Parks Associates, a research firm in Texas that specialises in the digital home (and which organised the conference at which Mr Burke gave his keynote address) surveyed a group of internet users and found that 84% of them use their PCs to store digital photos, 59% to store music, 36% for video clips and 26% for personal videos. If one includes devices other than PCs—such as TiVo, a popular digital video recorder— 17% also store movies and TV shows. In theory, these people could soon avail themselves of new wireless-networking technologies, such as an emerging standard called “ultrawideband”, to pipe all this content from their collections to electronic picture frames, screens and portable devices. Joined-up thinking That is not at all what they want to do today, however. Another study by Parks Associates found that 89% of people with a home-computer network felt that the relatively modest goal of sharing internet access is its most important function,
  4. 4. with printer-sharing the second priority. Worse, 27% of people who bought network gear said that they ran into problems during configuration, leading many to call the help desk of their internet service provider (who may or may not be responsible for the problem) at an estimated annual cost of $1.4 billion to that industry. Even downloading entertainment, as opposed to buying it on discs, appears over-hyped. According to a study by the OECD, there were over 230 websites offering 1m tracks in America and Europe at the end of 2004. But these online sales accounted for less than 2% of total music revenues; even with fast growth, they are projected to rise only by 5-10% by 2008. All this points to a huge problem with the digital-home vision: the lack, among most consumers, of any sense of crisis about the status quo in entertainment. “We don't think many folks are looking for an electronic nerve centre in their homes,” says Pip Coburn, who runs Coburn Ventures, a technology-consulting and investment firm. After all, popping in a DVD, say, is so easy and works so well. By contrast, getting a digital home up and running promises several lost weekends of fiddling with manuals and settings, and hefty expenses in new gear. According to Mr Coburn's formula for evaluating new technologies, whereby adoption is a function of the users' sense of crisis (ie, motivation to change) outweighing their perceived pain of switching, the digital home ranks as a clear “loser”. This miscalculation—if that is what it is—by the large vendors stems from their history of catering to companies rather than people, says Pure Networks' Mr Dowling (who used to be at Intel and who hired some 40 of his 60 employees from Microsoft). During the information-technology boom, the industry sold its wares mostly to chief information officers or chief technology officers with big budgets. These are customers who tend to be receptive toward buying “solutions” rather than products, and often hire consultants such as IBM Global Services to pull together hardware and software from various vendors. But “consumers don't buy as an IT manager does,” says Mr Dowling. “They buy spur-of-the-moment and hodge podge; they buy things, not systems.” To the extent that the digital home is not a thing but a solution, he thinks, “the vendors are all fooling themselves.” The vendors, naturally, disagree vehemently. “When you ask customers what they want, they will never tell you. You have to show them first,” says Microsoft's Mr Mundie. That is why Microsoft has, since 1994, had an impressive (or, to some people, intimidating) mock digital home on its campus in Redmond, Washington
  5. 5. State, which it updates with the latest gadgets. Intel, NETGEAR, HP and most other self-respecting technology firms have similar mock-ups for display. There is, argues Motorola's Mr Burke, a huge “need to educate consumers about the value of a connected home and lifestyle.” Talking the same language Outside the controlled environment of a mock home or conference demonstration, however, educating consumers tends to backfire. That is because real-world digital homes usually do not work very well. The premise of the entire vision, remember, is that heterogeneous devices talk to one another and readily transfer content to wherever the consumer wants to access it. This requires compatibility —“interoperability” in the jargon—among vendors involved in two technological categories. The first is file formats and codecs (short for coder-decoders), which encode digital information—such as a picture, song or film—compress it for transmission and storage, and decompress it again for viewing and listening. The second is digital-rights management software, or DRM, which protects such content against piracy and unauthorised copying. DRM allows the copyright holders of content— film studios and record companies, in essence—to define such parameters as when a film or song that is downloaded “expires”, or how many times it can be copied to another device, such as a portable player. The trouble starts here, with a bewildering list of acronyms that no ordinary consumer should ever have to know, but currently needs to know, to set up a digital home. The Moving Picture Experts Group (MPEG) is an industry body that defines widely used codecs such as MPEG-2 for video and MP3 for audio. But the big vendors prefer their own codecs—Microsoft its WM9 (short for Windows- Media-9), Apple, the market leader in online music sales, its AAC, and so on. In DRM, the situation is even more chaotic. Microsoft pushes its Windows DRM; RealNetworks, which makes rival media software, has Helix; Sony has OpenMG; Apple likes FairPlay, and so on. The upshot is that consumers cannot mix online services, gadgets and software from different vendors and be sure that the content they have paid for actually works. Music bought online from Microsoft's
  6. 6. MSN or Yahoo!, for instance, does not work on Apple's iTunes or iPod, and vice versa. This challenge is daunting because DRM technologies should not only be compatible today, but for all eternity. Otherwise, consumers will be afraid to pay for content, and will stick with CDs and DVDs, which seem painless and safe by comparison. “If consumers even know there's a DRM, what it is, and how it works, we've already failed,” says Peter Lee, an executive at Disney. The same goes for codecs. “The user shouldn't know or care what format they're using,” says James Poder, an engineer at Comcast, America's largest cable company and broadband internet service provider, because “consumers don't want to be IT administrators for their own home.” Prisoner's dilemma It may seem ironic, therefore, that vendors are refusing to make their technologies interoperable, thus potentially killing their own vision. On the other hand, it makes sense for each to try to make its own proprietary technology the winner, in order later to grab a disproportionate share of the market. The starting point of cable and telecoms companies, for instance, is as providers of broadband pipes into the home. So they are investing in IPTV (internet-protocol television), a vision in which content resides on the network and is pulled into the home on demand. Thus, says Cyrus Mewawalla, an analyst at Westhall Capital, a broker in London, America's Verizon and SBC and others are investing hugely in laying fibre-optic cables to homes (at a cost of about $1,000 per household), hoping that IPTV and the necessary set-top box could “evolve into the primary gateway to the digital home.” By controlling this gateway, they could offer a bundle of telephony, internet and entertainment, in effect “owning” the customer. This would at the same time help them to parry their biggest threat: Microsoft. Microsoft has itself invested in IPTV, ostensibly in partnership with telecoms and cable companies. Like its loss-making investment in game consoles (called Xbox), however, Microsoft intends this as a purely defensive hedge, says Matt Rosoff, an analyst at Directions on Microsoft, an independent research outfit near Seattle. Instead, thinks Mr Rosoff, Microsoft's strategy is to establish the Windows-run PC as the uncontested hub of the digital home. Hence its all-out push to establish its codecs and DRM as the standard. This would allow Microsoft to keep selling
  7. 7. Windows upgrades and to earn royalties from hardware and from consumer- electronics companies that make “spokes” for the Windows hub, such as portable music and video players, screens and online services. Microsoft's most explicit attempt so far is a version of its current operating system called Windows Media Centre Edition (MCE), which puts a simplified menu on top of the desktop screen for use with a remote control from the sofa. The MCE was first launched in October 2002, and has been upgraded several times since, but it has so far been mostly a dud, running fewer than 1% of all PCs sold last year. Microsoft now hopes to make MCE more relevant by selling “extenders”, little devices that can hook on to a TV set or stereo and communicate with the PC over a wireless network. Its biggest hope, however, is for Vista (previously known by the code name of Longhorn), the next version of Windows, which is due to be released late next year (after several delays). According to Microsoft's Mr Mundie, there is no question that the Windows PC will win this fight to become the central repository for all digital content, for a simple reason. The cable and telecoms companies, he says, are hampered by their business model, in which the set-top boxes sit on their own balance sheet and are leased, at subsidised rates, to consumers. This means that their incentive will always be to make the boxes cheaper. By contrast, Microsoft's incentive is to make its operating system more sophisticated, in everything from parental controls to usability. By the same logic, Microsoft will beat the consumer- electronics companies (such as Sony and Samsung). Their business model relies on selling devices rather than on recurring licence revenues. This leads to clutter in the home, without organisation of the content. Tom Berquist, an industry analyst at Citigroup, broadly agrees that the PC is likely to win. The on-demand world on offer from, say, Comcast, is simply not portable enough, he thinks. By contrast, he says, moving content to PCs potentially “liberates you from proprietary technology and lets you use content on any device.” In this sense, the only real competition to Microsoft is Apple, whose Macintosh operating system is widely considered to be more elegant and user- friendly than Windows, and which has a considerable headstart with the huge popularity of its iTunes music service and iPod player. Apple's problem, however, is that it has only 2.6% of the world market for PCs, whereas Windows runs on almost all the rest. Apple also differs from Microsoft in
  8. 8. that it simultaneously wants to be the main portable-device maker. It is, in other words, a software, hardware and consumer-electronics company all at once, and that does not leave much room for alliances with other industries to manufacture spokes for an Apple hub. There are signs that Apple is becoming more agnostic in order to compete with Microsoft. It has a deal with HP, traditionally a Microsoft ally (HP was, for instance, the first computer maker to ship Windows Media Centre Edition), under which HP bundles Apple's iTunes software on to PCs running Microsoft Windows. In a surprising announcement in June, Apple also said that it would start using microprocessors from Intel, another traditional Microsoft ally. Winner takes all? For the foreseeable future, the only certainty is that all these mighty companies will continue to preach interoperability while pursuing proprietary hegemony. This could lead to several scenarios. One is that one company, or camp, wins. The digital home, unified by the winner's standards, might then become a reality in the mass market. For this to happen, however, several companies and industries would first have to make huge strategic mistakes, and consumers would have to accede, in effect, to a repeat of the “Wintel” (Windows and Intel) near monopoly in the PC industry today. Another possibility is that the technology wars end with a truce, perhaps brokered by industry consortia that push open standards. This would be infinitely preferable for consumers and would probably make the digital home a reality much sooner, since it would mean that consumers could shop incrementally for new gadgets, all of which will fit with the others. The catch for providers is that this is much less exciting for their own bottom lines. There is a third possibility. This is that the wars continue, but consumers continue not to care. As John Barrett, research director at Parks Associates, says, “it seems that we've concocted a new variant of the ‘paperless' office.” This, you recall, was the consensus a decade or so ago among technophiles (but almost nobody else), that computer technology would save our forests by freeing us from having to read and write on paper. Today's variant, says Mr Barrett, is “no more tapes, CDs, DVDs, discs.” In other words, expect them to be around for a very long time to come.
  9. 9. The digital home Life in the vault Jul 1st 2004 | SAN FRANCISCO From The Economist print edition Companies are fighting to turn your home into an entertainment multiplex ON JUNE 21st, Intel, the world's largest chipmaker, launched two new lines of chips, code-named Grantsdale and Alderwood, which it says are the “most compelling” changes to the way personal computers (PCs) work in “over a decade”. From here on, claims Intel, PCs will be “all-in-one hi-fi devices”, “entertainment PCs”, and “vaults” for digital content. Intel's vision is that consumers will start to use their PCs at home to download, store and manage films, songs and games, in order to transmit all this fun stuff wirelessly to TV screens and stereo speakers throughout the house. The kids could then watch “Shrek 2” in the basement, while mum listens to Brahms in the kitchen and dad browses the holiday pictures on the main TV screen in the living room. As such, Intel's vision is neither new nor overly ambitious. For years, futurists have been peddling notions of digital nirvana in the home. In its wilder forms, this includes fridges that know automatically when to re-order milk via the internet, garage doors that open by themselves as the car approaches, and toilet seats that warm up at just the right moment. Most of this is guff. Nobody, aside from the self-selected early adopters at trade shows, would consider “upgrading” a garage door every few years for the latest
  10. 10. release. “In 20 years, my PC will still not be talking to my fridge,” says Jen-Hsun Huang, the chief executive of NVIDIA, the world's largest maker of graphics chips. On the other hand, says Mr Huang, “the vision of digital content is a much more compelling one than that of home automation.” And this is why the new chips may turn out to be as important as Intel claims. They are an opening salvo in a battle between the computer and the consumer-electronics industries over who will dominate the digital household. Intel, with a virtual monopoly in the chips that power PCs, naturally hopes that PCs will dominate and morph into “media hubs”. So does Microsoft, with its near- monopoly on PC operating systems. HP, Gateway, Dell and Apple also want the PC to win, although HP is also big in printers, digital cameras and other consumer gizmos, and Apple has the iPod to fall back on. On the other side are the giants of consumer electronics. Sony wants future versions of its game consoles, rather than PCs, to play the role of digital “hub”. TiVo, a leading maker of digital personal video recorders (PVRs), has hopes for its machines. So do makers of TV set-top boxes. Extrapolating from history, the PC industry would be the favourite to win, since it has the powerful and rich Microsoft on its side. Microsoft is certainly trying. In the past year, it has launched and re-launched its Windows Media Center, a version of its operating system that looks more like a TV menu and can work via a remote control. Microsoft is also pushing its own next-generation DVD technology, that competes with rival technologies from Japan's Matsushita, NEC, Toshiba and others. Microsoft's problem is that consumers do not seem keen. Only 32% of American households with internet access polled by Parks Associates, a consumer- technology consultancy, said they were “comfortable” with their PC becoming an entertainment system. Nobody wants to watch the system reboot during a good movie. This scepticism, however, does not automatically mean that the consumer- electronics industry will win. The one thing that all companies seem to agree on is that households will be connected to the internet via a broadband link that is always on, and that content will be shared wirelessly between rooms within the
  11. 11. home. The upshot is that there need not be any single device inside the home that becomes a central media hub. A baby picture could be stored on a PC, on a console, or on a mobile-phone handset. Or it might alternatively be kept on a remote and powerful “server” computer somewhere on the internet. The latter model is how subscribers to Rhapsody, a service provided by RealNetworks, an internet media firm, already listen to music. The gadget makers therefore have much to ponder. Art Peck, an analyst at Boston Consulting Group, says that the real money in the digital home will be made by those providing a service or selling advertising. The hardware makers, he thinks, are therefore fooling themselves by thinking that any device can become a “Trojan horse” to enable them to capture the bounty. It is much more likely that they will all end up as makers of interchangeable commodities for the digital home, that the consumer cares little about unless the stuff breaks down The digital home Boxed in Nov 24th 2005 | SAN FRANCISCO From The Economist print edition Big technology vendors are piling into people's living rooms IT HAS been quite a week for the “digital home” that so excites technology vendors. On November 18th, Cisco, the world's largest maker of the routers and switches that direct internet traffic, said that it would pay $6.9 billion to buy Scientific-Atlanta, the second-largest maker (after Motorola) of the television set- top boxes that unscramble the signals of cable operators. Then SBC became America's largest telecoms company by completing the acquisition of its former parent, AT&T, promptly giving itself a lower-case “at&t” logo and reaffirming its strategy to transform itself into a provider of on-demand video services, delivered over internet links. TiVo, a maker of digital video recorders, said that it would add
  12. 12. software to let users transfer saved TV shows on to iPods. And Microsoft announced a new version of its Xbox game console (see article). Behind all these events is the widespread belief among technology companies that consumers, rather than companies, are their best remaining hope of returning to the revenue growth seen during the dotcom boom of the 1990s. The consensus is that consumers will use one pipe—their internet connection—not only for their web surfing and telephony but also for their films; and that, having pulled their entertainment from the internet, consumers will then beam it to any screen or room in the house over supposedly “seamless” wireless home networks. This, in marketing circles, is called the “digital” or “connected” home. Behind these hopes lies a lot of desperation. Telecoms firms are in trouble because their fixed-line business is shrinking, and because their cable rivals are beating them in entertainment while getting stronger in telephony. Like telecoms firms elsewhere in the world, AT&T and its arch-rival Verizon are fighting back by laying new fibre-optic cables in the ground so that they can provide the bandwidth necessary to carry television over internet protocol (or “IPTV”). Their goal is to provide a “triple-play” bundle of data, voice and video (or even a “quadruple-play” including mobile-phone service)—presumably so that customers can never escape them again. Hence Cisco's attempts to follow the service providers and break out from the bowels of the internet into the plain sight of living rooms. In 2003 Cisco bought Linksys, a firm that makes gadgets for home computer networks. The deal with Scientific-Atlanta is a natural next move. The idea is that Cisco's gadgets will become the end-point of the internet-video services offered by the telecoms and cable companies. Once video from an IPTV service has been saved on to the hard drive of a set-top box, the same gadget could use Linksys technology to beam the video to speakers and screens in other rooms. The cliché for all this is “convergence”—the folding of several separate services (voice, television, data) into a single stream of zeros and ones, sent using internet protocol. But “collision” is probably a better description, says Pip Coburn of Coburn Ventures, an investment adviser. “The vendors are all scrambling and steadily butting heads more and more,” he says. “Cisco is inches away from a major battle with Microsoft now.”
  13. 13. Not officially, mind you. Microsoft has tactical or defensive stakes in all parts of the digital home, so it has partnerships with the telecoms companies and with Cisco. But Microsoft is really hoping that the “hub” of tomorrow's digital homes will be a computer running the “media-centre edition” of Windows (in effect, Windows with a remote control). As second-best, Microsoft would like its Xbox console to become the control centre of the digital home. In short, Microsoft wants anything but the boxes offered by other companies. Then, of course, there are Apple, Sony, Motorola and Yahoo! Each of these firms is also trying to secure a lucrative place in the digital home. The result, for the time being, is a tug of war among incompatible digital-copyright codes that leaves consumers, who simply want things to work, confused or indifferent. The digital home is likely to remain a dysfunctional one. Wi-Pie in the sky? Mar 9th 2006 From The Economist print edition Communications: Cities across America plan to build municipal Wi-Fi networks to widen access to broadband. Will they work?
  14. 14. “WE WILL not stop until every San Franciscan has access to free wireless-internet service.” It was a typically bold statement from Gavin Newsom, the charismatic young mayor of San Francisco, as he announced plans in October 2004 for a Wi-Fi network that would blanket the city with wireless-internet coverage. Mr Newsom thus joined a nationwide movement of cities across America that are planning to provide wireless-broadband access for government workers, residents and businesses. These municipalities, ranging in scale from communities of a few thousand residents to huge cities including Philadelphia and San Francisco, are concerned that the lack of availability of broadband access, compared with other parts of the developed world, is holding back economic growth and perpetuating a “digital divide” between internet haves and have-nots. A further motivation is that by setting up their own wireless networks, municipalities hope to be able to cut communications costs, improve the efficiency of their staff, and make possible new services such as allowing parking meters to accept debit and credit cards. Small municipal wireless networks, typically built for local-government use, have been up and running in some parts of America for some time. The far bolder idea of building citywide networks available to all took flight in August 2004, when plans for such a network were announced by John Street, the mayor of Philadelphia. Stringing transmitters across the entire city would create the world's largest Wi-Fi hotspot, providing access both indoors and out. This would extend low-cost broadband access to existing users frustrated by the slow speed and high cost of dial-up internet connections. A survey conducted by the city found that 72% of internet-connected households used dial-up connections, compared with 47% nationally. “We wanted broadband at dial-up rates,” says Dianah Neff, the city's chief information officer. In addition, the city would provide subsidised or free access to tens of thousands of unconnected residents, sometimes even supplying computers. Only 45% of Philadelphia residents had internet access at home, the city found, which compared poorly with the national figure of 73%. Last October the body established by the city to oversee the project, Wireless Philadelphia, picked EarthLink, a national internet service provider, to build and operate the network, which is expected to cost around $10m.
  15. 15. Philadelphia's ambitious scheme prompted many other cities, including San Francisco, Portland and Minneapolis, to follow. So far, nearly 200 municipalities have announced plans for citywide wireless networks, issued bid requests, or built networks, according to Esme Vos, the founder of, a website that tracks the subject. Over the next three years nearly $700m will be spent building such networks in America, she estimates. Some networks will be supported by advertising; many will charge fees of $15-25 per month. Most will offer some form of free access at certain times of day or to poorer users. Free (or at least cheap) broadband for every citizen—who could argue with that? Plenty of people, it turns out. Critics worry that cities are underestimating the cost and complexity of building and running their own networks. Incumbent telecoms and cable operators are predictably opposed to subsidised schemes that will compete with their own broadband offerings. Then there are the technical objections: no Wi-Fi network as large, dense and complex as Philadelphia's proposed system has ever been built; citywide Wi-Fi networks could interfere with existing Wi-Fi systems; and the networks will be built using proprietary technologies, so that municipalities will become dependent on their equipment- makers. These criticisms have some merit. Last October Mr Newsom back-pedalled on his “free Wi-Fi for all” pledge, in favour of providing “affordable” wireless access throughout San Francisco. And when putting networks out to tender, the preferred model is now for the “Using Wi-Fi at construction and operating costs of the network to be high power levels for citywide carried by the winning bidder, to ensure that cities' coverage could own liabilities are limited. But much uncertainty still drown out remains over whether large-scale municipal Wi-Fi existing networks can be made to work at all. Even if they can, networks.” nobody knows how much they will cost. And it is hardly reassuring that rival manufacturers, each promoting different and incompatible technologies, claim not just that their equipment is the best, but that their rivals' will not work at all. When challenged, vendors point to the many smaller networks that are already running. So who is right? The view from Tempe
  16. 16. To see municipal wireless networking in action, a good place to start is the city of Tempe, Arizona, a suburb on the outskirts of Phoenix with 160,000 residents spread over 40 square miles. The flat, desert terrain is an ideal environment in which to use wireless networking to boost the availability of broadband. Several years ago the city installed wireless-broadband equipment on two buttes at the opposite ends of town. Rather than paying to lease high- speed digital lines from Qwest, each of which would cost hundreds of dollars per month, police stations and other municipal buildings are linked via the city's own wireless network. Telecoms costs have fallen from $1.7m a year a decade ago to $0.5m today. “We're probably not Qwest's favourite customer,” says Dave Heck, Tempe's deputy manager of information technology. As a result, when the city began considering ways to extend broadband access to more residents—including nearly 60,000 students, staff and faculty members at the main Arizona State University campus in the centre of town—wireless made a lot of sense, says Mr Heck. “We don't have a lot of competition for broadband in Tempe,” he says. “There's just not a lot there.” A local provider, NeoReach, won the contract to build a Wi-Fi “mesh” network to provide broadband throughout Tempe, using equipment made by Strix Systems. NeoReach is paying for the network's construction, and will collect access fees from subscribers; the city's administration, police, fire and emergency services will also pay to use the network, which will cost $2.3m to build. Mesh networking allows large areas to be blanketed with wireless coverage quickly and inexpensively. As its name suggests, a mesh network consists of an array of wireless access points, only a few of which are actually connected back to
  17. 17. the internet via high-speed links (known as “backhaul” connections). The trick is that all of the access points double as relays, passing packets of data to and from their neighbours. This connects up the mesh, so that users can access the internet at high speed at any of the access points. If the nearest access point does not have a backhaul connection, the packets of data that users send and receive simply make one or more “hops” across the mesh. As well as being cheap and fast to set up—partly because many of the access points can be attached to utility poles—mesh networks have several other merits. They can provide coverage in areas, such as sprawling suburbs, where fast copper or fibre-optic connections are hard to come by. “When you get out in the residential areas, there's no fibre,” says Chuck Haas of MetroFi, whose company has installed mesh Wi-Fi networks in three of the San Francisco Bay Area's largest suburbs. Mesh networks are reliable, since the failure of one or more access points does not bring down the whole network, and they can also route data around obstacles, such as large buildings, which might otherwise block coverage. By January NeoReach had installed over 400 access points in Tempe, providing Wi-Fi to subscribers in much of the city, and in outdoor areas in downtown Tempe and on the university campus. With the network's completion this month, NeoReach claims to have built the first example of a citywide Wi-Fi network in America. Proof, surely, that Wi-Fi mesh networks are an excellent way to extend broadband coverage across entire cities? In fact, it is too soon to say. Wi-Fi was designed for use in small hotspots, not in citywide mesh networks. Using mesh networks to connect up a flat, sprawling, relatively sparsely populated suburb is one thing. But many unknowns remain, given the scale and nature of the networks about to be installed in Philadelphia, San Francisco and several other large American cities—all of which will push Wi-Fi technology to its limits. “We're taking this technology and, through some enhancements, we're trying to apply it for an application it wasn't really designed for,” says Peter Rysavy, an independent wireless consultant. “Consequently, we just do not know how well it's going to work—but we're going to find out.”
  18. 18. Not so fast Some mesh network devices, notably those made by Tropos Networks, the market leader, use the same Wi-Fi frequencies both for communication between neighbouring access points, and to connect users to the mesh. Wireless experts fear that the use of Wi-Fi at the very high power levels required for municipal- scale networks, particularly in cluttered cities, could drown out existing outdoor Wi-Fi networks, public and private, and slow or disable indoor networks. An estimated 10m American homes contain Wi-Fi networks; many more are installed in offices, schools, coffee shops and parks. “It is absolutely guaranteed that you're going to end up interfering with anybody else that is using any channels within your coverage area,” says Mr Rysavy. To make matters worse, Wi-Fi operates in “unlicensed” frequency bands, which are supposed to be available for anyone to use. By interfering with existing Wi-Fi networks, a municipal network is, in effect, appropriating a shared public asset. “There are going to be 3,000 or 10,000 or whatever access points in the city of San Francisco,” says Tim Pozar, a wireless consultant. “Even though this is unlicensed frequency, guess who owns the spectrum? The vendor.” Ron Sege, the boss of Tropos, insists that his firm's technology does not cause undue interference. After all, he notes, there are millions of access points in use already, with more being added all the time. Tropos nodes installed in a city are no different, he says, from a home user buying an access point and plugging it in. The problem, however, is that off-the-shelf access points operate at just 3-20% of the power output of the mesh nodes made by Tropos and its rivals. Reducing the signal strength of the mesh nodes would reduce interference with existing networks, but the density of nodes would then have to go up, increasing costs. Interference is not the only potential difficulty with municipal Wi-Fi networks. Another problem, notes Mr Rysavy, is that there is no common standard for Wi-Fi meshing, and thus no compatibility between the five leading vendors' equipment. So if a city builds a network and the vendor goes bust—as recently happened to Vivato, a pioneer of long-range Wi-Fi gear—keeping the network running could be difficult. Although a standard for Wi-Fi meshing, called 802.11s, is under development, it is not very far along—and cities want to start building now. “Until there's some standardisation in the mesh protocols, any deployment today is pretty risky,” says Mr Rysavy.
  19. 19. That raises another criticism: that wireless technology is developing very quickly, yet most municipal networks will be based on a Wi-Fi standard that is already three years old. Why not wait for the next-generation standard, 802.11n, which is much faster, or for Wi-Fi's long-range big brother, a much-hyped technology called WiMax? Waiting a year or two might make it possible to build faster networks over greater areas at lower cost. In fact, 802.11n is designed to perform best indoors, while WiMax has already been factored into most municipal wireless plans—where it will be used to provide backhaul connections to the mesh. EarthLink, for example, will use Canopy, Motorola's WiMax-like technology, for this purpose in Philadelphia and in four other cities. And dedicated Canopy links will also, says Ms Neff, be used to connect several hundred large buildings, which will not therefore rely on the Wi-Fi mesh. “Metro Wi-Fi is going to create the base and the excuse for WiMax,” says Tom Hulsebosch of Motorola's Canopy division. Municipal Wi-Fi must also contend with political interference. A number of reports financed directly or indirectly by incumbent telecoms firms such as Verizon, AT&T and Comcast, have criticised municipal networks as risky wastes of taxpayers' money that will provide unfair competition to private companies. Similarly, there have been complaints that if city governments cannot even fill potholes in roads, how will they be able to manage the far more complex task of running a network? Proponents of municipal networks respond that broadband access is just another utility, and that utilities have been state-run in the past. That is true, but public ownership of electrical utilities was, in fact, as controversial a century ago as municipal Wi-Fi is today. Philadelphia's plans were almost derailed when, after “If city much lobbying, legislation was proposed in governments Pennsylvania that would have given Verizon, the local cannot even fill incumbent, right of first refusal in any municipal potholes, how will telecoms scheme. A bitter argument ensued, and the they be able to manage running a bill was signed into law only after Verizon agreed to network?” issue a waiver exempting the city of Philadelphia from the provision. Similar bills preventing municipalities from going into competition with telecoms incumbents have been passed in over a dozen states.
  20. 20. One exception is Texas, where a grass-roots movement succeeded in scuttling a vote on such a bill, which had been sponsored by the local incumbent, SBC (since renamed AT&T). Michael Dell, the founder of the world's biggest PC-maker, based in Round Rock, Texas, helped to undermine the bill when he pointed out to state legislators that more broadband might sell more computers, and thus bring more jobs and tax revenues to the state. To avoid such political wrangling, many cities are now taking the approach of awarding contracts to firms that build the networks at their own expense, pay taxes or franchise fees, and operate autonomously, in return for special rights to city-owned facilities such as utility poles, towers, building tops and electricity. All of this means that Philadelphia, San Francisco, Minneapolis, Portland, Chicago and many other cities are participants in a great experiment. Building a large Wi- Fi network is harder than it looks, says Mr Haas, who has first-hand experience, not least because, once deployed in the field, mesh Wi-Fi equipment does not always perform as promised. “The radios we put up in 2002 and 2003 and the first half of 2004—hundreds of them—are either back at the vendors or sitting in the warehouse,” he says. Only in the past 18 months has his firm hit upon the right combination of equipment and network design. One step at a time Reassuringly, Philadelphia and other big cities are taking a staged approach to the deployment, which will allow the conflicting claims made by equipment manufacturers to be resolved, one way or another. In Philadelphia, EarthLink will start by building a 15-square-mile test network covering different terrain, demographics and building densities. Building such a test network “is very expensive to do, but it's never been done,” says Greg Richardson of Civitium, a consultancy that was involved in the bidding in Philadelphia and San Francisco. If EarthLink cannot meet the specifications set by the city and assessed by independent auditors, the network will not be built. “It gives us a chance to see if what we ask for is realistic, or if we need to change some of the parameters,” says Ms Neff. And if existing Wi-Fi networks are disrupted, citizens in the test area will know whom to blame.
  21. 21. Mayor Newsom's utopian vision of free wireless for all quickly ran up against the snag that all such giveaways require: there's a price to pay at the end of the day. Even with the decision to charge some users for access, not all cities will find that their plans for citywide Wi-Fi networks turn out as expected. The real measure of municipal wireless networks will not be in places such as Tempe, where they are expected to work, but in bigger cities, where success is far from certain. Whether service providers will be able to meet the required technical standards and still make a profit will soon become clear.