Wi-Fi is a big story. We read about how Cometa, Verizon, T-Mobile, Wayport and Boingo are rolling out and aggregating wi-fi hot spots by the thousands, how broadband customers are using wireless access points as cordless phones for their laptops, and how wireless functionality in laptops is moving from interface cards to CPUs.
Credit, as usual, goes to the big vendors, the chip makers, the aggregators, the supply side of the market in general. Meanwhile, as usual, the leading edge is being moved forward by hackers and their friends at the grass roots level -- just as it was with the Net and with Linux (and still is, in both cases).
And the work that's likely to do the most good for the most people -- for civilization itself -- is happening in public, outdoors, on the streets and parks of towns and cities.
Public wi-fi is hacker community outreach. For hackers, it's a way to bring broadband Internet to public spaces. For home broadband Internet customers, it's a public way to share private bandwidth. For users in streets and parks, it's a more civilized public life. Unwired from providers, public wi-fi makes the Net a gift: a civic grace akin to parks, sidewalks, boulevards and libraries.
This past May, when the FCC continued deregulating ownership of what we used to call the "public" airwaves, the agency made a big deal about "saving" what was left of "free over-the-air" broadcasting. The Internet, however, needs no deregulation -- or regulation -- to make it free over the air. All it needs is generous technologists, citizens and civic organizations.
That's what we have in New York City today. And their work is emarkable to behold.
New York unwired
Some public wi-fi efforts are largely municipal. That's the case with Long Beach CA, which provides a large public "hot zone" in its downtown and another at its airport.
Other efforts are driven by technically savvy volunteers. That's the case with Austin, Portland, Seattle, San Francisco, Winston-Salem, Seaside and Greenwich -- among many other places. Outside the U.S., Consume.net in London and WAfreenet in Perth are both good examples of volunteer groups at work.
Independent private concerns are also doing their part. In Asheville NC, for example, Britt Blaser recently discovered a hydrant-sized free public hot spot. Called a BeamPost, it also features ethernet, infrared and Bluetooth ports. The Asheville newspaper says BeamPost is the product of a Natural Communications, a public-spirited private company.
New York, however is different breed. It's all the above.
While New York ranks #27 on Intel's list of "most unwired" cities (Portland is #1), it is perhaps the #1 example of working consensus -- between hackers, businesses, government entities and various nonprofits -- about the need for free public wi-fi. This consensus is what gave rise to NYCwireless (www.nycwireless.net), a self-described "loose collection of interested minds." NYCwireless has two missions (quoted from the Web site):
New York is one of the few places in the U.S. where public and private interests are generally considered symbiotic. Rather than the public enemy we hear about on talk radio, government in New York is widely regarded as a provider of the public infrastructure necessary to support private enterprise. NYCwireless is proof in the pudding.
The founders of NYCwireless are Anthony Townsend and Terry Schmidt, partners in Emenity, the company that has been building out the new ad hoc NYCwireless infrastructure in New York. Both NYCwireless and Emenity are products of public and private symbiosis. Same with its customers, which include publicly funded neighborhood associations created for the purpose, among other things, of building out infrastructural improvements -- such as public wi-fi in the parks.
In May, New York's City Council issued a staff report titled Network NYC: Building the Broadband City. It grew out of a hearing last December of the Select Committee on Technology in Government and the Subcommittee on Small Business, Retail and Emerging Technologies. It recommends a restructuring of the city's fractured broadband procurement methods, a new fiber/wireless metropolitan area network (MAN), and public wi-fi networks. As an example of the latter, it says a potential Prospect Park wi-fi network would cost $192,000 to create, but little to maintain. The cost of bandwidth, provided by the city's new fiber backbone, would be zero.
That report opens with special thanks to Anthony Townsend, who is also a Research Scientist at NYU's Taub Urban Research Center, where he has produced a pile of wise and seminal papers (http://www.informationcity.org/research/) about the growth of the Internet in urban settings. With NYCwireless and Emenity, Townsend has taken matters into his own hands.
Terry Schmidt's job is turning Anthony's vision into reality. Terry is Emenity's CTO and the hacker behind the Pebble Linux distro, the stripped-down Debian used by NYCwireless access points. Pebble is freely available to anyone else who wants to build out the same kind of thing. (See Kurt Starsinic's sidebar detailing Pebble.)
Going signal fishing
Wi-fi range is low, on purpose. It operates on a tiny wedge of unlicensed frequencies -- just 14 channels between 2.412 and 2.484GHz. (Here in the U.S. we only use 1-11. In Europe they use 1-13 except for France, where they use 10-13. Japan runs from 1-14.) The default transmission power of most access points (also known as APs, WAPs, hot spots and base stations) is 30mw, or 3/100 of one watt. That's about 1/10 the power of your basic cell phone, but on a higher frequency, where the energy attenuates more rapidly with distance through air, and has trouble penetrating opaque objects. (Loss through glass runs around 3db, while some tinted glass acts like shielding.)
Between output power and the inverse square law (by which signal strength declines asymptotically with distance from the source), wi-fi range tends to run less than the average cordless phone (which also sports a more powerful signal).
With such a handy service delivered by such a short-range signal, it's only natural to find the best signal fishing where the population is both dense and conveniently arranged -- such as New York City, where people live and work on top of one another and where broadband access is easily available.
I had my New York wi-fi epiphany this May, when I stayed at my friend Britt Blaser's apartment in Manhattan (Britt is an open source project called Xpertweb.) When I opened my laptop, I saw a pile of signals, in addition to the one from my friend's home access point. I also heard a lot of buzz about NYCwireless, the voluntary organization that was doing a great job of delivering free wi-fi to parks and streets all over town.
After reporting my discoveries to Don Marti (our Editor-in-Chief), we decided it would be a cool idea for me to come back and take a closer look at what's going, on from a Linux angle.
So I spent six days in Manhattan in late May, touring its wi-fi-enabled parks and cafes with hacker docents, meeting with movement leaders, wardriving* around in taxis and otherwise having a great time discovering that there's even more going on than we had suspected.
Driven by distraction
My primary instruments were two laptops kindly loaned by their makers: a Lindows MobilePC (running Lindows' version of Debian) and an Apple 17" Powerbook. The Lindows unit was equipped with a NetGate 200mw card loaned by friends at Vivato. The Powerbook had an internal card with an antenna embedded on the edge of its lid. We also made use of Kurt Starsinic's laptop, which runs BSD.
The ideal way to go signal fishing with a Linux or BSD laptop is with Kismet, a wireless network sniffer so full-featured it even does neat stuff with GPS, precisely associating signals with locations. Unfortunately, we weren't able to get the drivers that would make Kismet work with the card (which is too bad, since the performance of the laptop/card combo was impressive), so we relied on the Powerbook and Macstumbler, an open source tool that lacks Kismet's nifty extra functions but still does the basics. It provided more than enough data to crunch.
I ran six sessions, which included nine taxi rides, nearly all in Manhattan (the last session ended with a highway ride to LaGuardia Airport through signal-free parts Queens). Each session recorded basic data about every detected wi-fi signal, including SSID (Service Set Identifier - the signal's name), MAC (Media Access Control - the hardware fingerprint) address, WEP (Wired Equivalent Privacy) status, channel, maximum signal and vendor, among other things.
Here's a summary the findings:
|Session||Date||# of APs|
|Total unique MAC addresses||1112|
|SSIDs (those with more than 10)|
On city streets in Manhattan, I found there was nearly always an access point in range. And I'm sure the numbers above would have been much higher if I'd had an antenna outside each taxi instead of on my lap in the back seat. While there are lots of commercial hot spots, the vast majority of them appear to belong to individuals.
The willingness of individuals to share bandwidth is amazing, considering the security risks involved. Leaving an AP open is kind of like leaving the front door unlocked with a post-it note inviting people to come in and raid the fridge (the risk is to your data, not your bandwidth, most of which you're not using, most of the time). While the number of wide-open APs was lower than the WEP numbers above suggest (quite a few were password protected), there were still plenty of usable signals. More than once I was able to pick up and send email while a cab was stopped at a light.
I based my visit out of Britt Blaser's aprartment, not far below the penthouse level of a high-rise at the corner of 43rd Street and 2nd Avenue in midtown Manhattan.
Britt's apartment has one of the best views in town. To the East, it squarely faces the United Nations building, which looms over the East River like one of those upright slabs from the movie "2001". To the UN's south is Tudor City, an ornate apartment complex built in the 1920s. Remember the Gothic rooftop home of Norman Osborne (played by Willem Dafoe) in Spider-man? That location is one of four similar "mansions" (each a clever cover for water tanks, elevator pulleys and other rooftop uglies) that sit atop Tudor City's apartment buildings like castles on mountains. Britt's place looks down on all of them, plus the pretty little park in their midst. To the West are two of New York's tallest skyscrapers: the Chrysler Building and Conde Nast Building, labeled 4 Times Square.
Height has value in New York. That goes for high frequency broadcasting even more than for real estate. All the highest buildings bristle with antennas. Topping the beautiful old Chanin Building nearby on 42nd street are two FM broadcast antennas: an ancient six-bay mast of horizontal clover-leafs that once radiated WQXR, the Classical FM station of the New York Times; and a new two-bay antenna that is probably the auxiliary for WQXR or some other station.
The need for auxilliary transmitter sites was brought home when nearly all of New York's TV and a third of its FM stations were destroyed in the 9/11 attack. All of them radiated from one big mast on the North Tower. It was the last piece to hit the ground.
The problem still isn't solved. Major construction was happening over on the top of the Conde Nast building, where workers with cranes were raising the new broadcast tower that will carry the signals of FM stations whose transmitters were lost nearly three years ago. The project is taking a long time because the Empire State Building is already overburdened with TV and FM antennas (including new digital HDTV antennas, which were also lost on 9/11). No engineer wants his station's signal shadowed by higher buildings.
The gargantuan cost and effort required to continue serving brute-force analog radio to increasingly disinterested citizens provided an ironic contrast to the grass-roots wi-fi movement carried out by those citizens in the streets and parks below.
As an old broadcast engineer and ham radio guy (from waaay back), I couldn't wait to take advantage of Britt's high-rise, to see if we could provide wi-fi service to Tudor City's park, half a block away. But the results were disappointing. Yes, we could deliver a barely-useful signal to one side of the park, but only when Britt hung his Linksys WAP-11 outside the window.
Thus began my acquaintence with the engineering facts of wi-fi life. I had to start by throwing much of what I knew about broadcasting out the window.
Just before the trip, I mentioned in a SuitWatch newsletter that I'd be coming to New York to check out the wi-fi situation, and that I could use some local help. The first reply came from Kurt Starsinic, who quickly became my wi-fi docent for warwalking and wardriving through lower Manhattan. (See Kurt's sidebar on the Pebble Linux distro).
I hooked up with Kurt at Alt.Coffee, on Avenue A across from Tompkins Square Park. Alt.Coffee is both a comfortably run-down coffee house and a reliquary for dead computers. Kaypros, ARCnet hubs, early-vintage PCs and other antiques (including a working 8-track player sitting under a monitor, emulating a cpu) are scattered on tables and piled up in corners. An artfully arranged graffiti-covered pile of dusty old PCs, hard drives and whatnot is fixed to the wall in the rest room. Worth a visit.
As it happened, the APs for both Alt.Coffee and NYCwireless were down while we were there; but when we walked around Tompkins Square Park, we still found at least one home node with an open and usable connection. (And yes, of course we used it.)
Our next stop was City Hall Park, where the NYCwireless signal is clear and strong. There I was able to sample NYCwireless' local fare (the splash screen carries lots of useful neighborhood information) while Kurt briefed me on technology issues and both of us waited for my old friend Stephen Lewis to show up.
Steve, who carries U.S. and Dutch passports, is a European telco industry veteran who was highly curious about what was happening with wi-fi in his home town. Walking around Steve's old haunts in the Lower East Side, we were impressed by the density of wi-fi, from both public (e.g. NYCwireless) and private sources: Verizon public phones, McDonalds restaurants and Starbucks coffee shops, in addition to private homes.
The ability to get on the Web almost anywhere in an outdoor urban setting was especially impressive to Steve, a two-time Fulbright Scholar with a hearty appetite for information. As a result he began to develop ambitious plans to carry the lessons of New York neighborhood wi-fi (including Linux technologies) to Bulgaria, where he has lived for much of the last decade. (See sidebar.)
One of the most interesting figures in the New York wi-fi movement is Drazen Pantic. A former mathematics professor at the University of Belgrade, Drazen ran the Internet service of B92, a radio station that was a thorn in the side of the Milosevic regime. After the station's transmitter was mysteriously shut down, Drazen made sure the station's news and information continued to come out on the station's Web site, and through streams that were picked up and rebroadcast in the U.K., Netherlands, the U.S., and, most significantly, Yugoslavia. Stations there picked up and rebroadcast the analog signals relayed by satellite from the Netherlands. As a result, B92 quickly became the primary source of news from, and about, Yugoslavia and the conflicts there. Hearing him tell the story of his life, it was clear that Drazen was a hero of several revolutions at once.
And more to come. We met several times during the trip, mostly over coffee at a cafe in Soho that's just up the street from Location One, a multimedia art gallery that also happens to be the U.S. base of the Open Source Streaming Alliance. Drazen is also involved with Dyne.org, a Vienna based group of free software hackers devoted to producing GPL'd freeware for realtime video processing, media streaming and other cool stuff. The coolest of Dyne's tools, Drazen explained, is HasciiCam, a neat little hack that captures video from a TV card, renders it into ASCII and outputs it in a variety of ways -- as HTML with a refresh tag, as a live ASCII window, or as a simple text file.
HasciiCam is made for the streets. The software will grab video using Dyne's Video4Linux API, render it into an ASCII representation using the AA-lib (a portable ASCII art GFX library) engine and wrap it in HTML. You can also use jpeglib to sequentially dump small .jpg images. The result is browsable without the need for plugins or java.
On the downstream side, Drazen is excited about both the Dyne:bolic Linux distro and MPEG4IP.Dyne:bolic is a multimedia-oriented distro that can run from a CD and recognize sound, video, TV, network cards and other peripherals. MPEG4IP is a streaming package that obviates the need to use proprietary streaming systems from Real, Microsoft and other large players that charge money to use their software. Drazen says, "After downloading Dynebolic, you can burn a CD, boot into Linux and stream high quality MPEG4 -- preferably using an OSSA streaming server."
As for hardware, Drazen invites interested readers to dig iTuner's Media Box, a very slick piece of Linux-based rack-mounted multimedia streaming hardware. And not expensive, considering.
The project that sums all this up is Wireless Broadcast Public WiFi Network 2 Cable Network. (http://open4all.info/laika) Here's his illustration:
Drazen believes all these open source efforts will finish liberating audio and video authoring, production and distribution from the corporate chokeholders that still hold our ambitions and imaginations in check. He sees Linux as the public OS platform, and wi-fi as the public network commons. Together they'll support a new form of (literally) public TV and radio. Between wi-fi, HasciiCam, digital camcorders, cheap hardware, free software, Dyne:bolic and MPEG4IP, Drazen expects the threshold of reporting and broadcasting to drop about as far as it can go. When it gets there, watch out.
Luggables and wearables
On Sunday I returned to Alt.Coffee to meet with Ahmi Wolf. He and Mark Argo are the creators of the Bass-Station, a turn-of-the-80s suitcase-size ghetto blaster that also happens to be a digital juke box and a wi-fi hot spot. Ahmi and Mark removed the radio and cassette components of this funky old thing, and replaced them with a variety of modern portable wi-fi goods: Viatech mini-itx motherboard, wireless interface card hooked to an antenna, Debian (Woody) loaded onto a compact flash card, and a 120 Gb hard drive. They left the amplifier and speakers, and hooked them up to the board's audio output.
The result is the social and aesthetic opposite of an iPod: a big ugly stereo that's also a Linux-based wi-fi access point, plus a juke box with a big-ass hard drive. (Ahmi's sidebar explains the whole hack in detail.)
The idea was to create a juke box for all kinds of convivial settings -- from parties in parks to hang-outs on college campuses. Everybody connected by wi-fi to the Bass-Station gets to contribute music and play disk jockey, so it rewards cooperation as well.
The social aspect is critical, because the Bass-Station belongs to the neighborhood extranet -- not to an individual, and not to the whole world. Ahmi explains:
The Bass-Station is not connected to or a part of another network. It creates it's own network that only exists within the range of the Bass-Station itself. On one hand the range of wi-fi is limited; but the limited range makes it special. Users of the network are all in close proximity to each other. That makes them members of a community -- be it a stable, persistent community or a spontaneous and mobile one like the Bass-Station's network.
I see Bass-Station as a prototype for a whole new consumer electronics category. Like HasciiCam, Bass-Station re-imagines an industry from the outside in. As with Linux and open source development, it's a way that the demand side gets to supply itself. Smart suppliers will get the clues.
Steve Lewis and Dave Pentecost showed up later at Alt.Coffee. Dave has been another valuable source of contacts and wisdom about wi-fi in New York. When Dave isn't on assignment for National Geographic or doing other interesting work in exotic locations, he's an urban homesteader and community activist in his Manhattan neighborhood. His Web journal, The Daily Glyph, is one of the best of its breed..
We all went to dinner at a Moroccan place around the corner, where we met up with a larger crowd that included Ahmi's Bulgarian friend Milena Iossifova, a fellow student at NYU's Interactive Telecommunications Program. (I was referred to both Ahmi and Milena by Clay Shirky, who teaches in the program. Ahmi, Mark and Milena have all been students of Clay's.)
Milena has a way-cool wi-fi creation of her own called WiFisense, which she calls "a wearable scanner for wireless networks." It's a silver reflective handbag with 64 LEDs in three different colors, each turned on by wi-fi activity on a different channel.
Steve and Milena immediately began talking in Bulgarian while everybody around the table shared a sense of participation in something truly revolutionary. The optimism and energy reminded me of what Silicon Valley felt like back in the 80s and 90s, but without the corrupting context of other people's money. It helped that Ahmi, Milena and Dave had already produced enabling goods in this new culture. They were part of something that was actually happening.
Building a new infrastructure, one Pebble at a time.
I met with Terry Schmidt at Emenity's offices near Wall Street, where he briefed me on the challenges of deploying public wi-fi in New York's peculiar urban settings.
The first big project for both NYCwireless and Emenity was Bryant Park, which shares a midtown block with the New York's Public Library..
"We overbuilt that one with two omni antennas, one sector antenna and two point-to-point links within the park itself. But it was a big success, so it became clear that there was a need for free wireless networks. A volunteer organization like NYCwireless can't easily do service level agreements and stuff like that, so that's what we provide with Emenity."
Terry sees Emenity as a midway organization between the purely voluntary and the purely self-reliant. Bryant Park, for example, was originally built by NYCwireless, then maintained by Emenity, and is now entirely run in-house by the park itself.
As an after-market business, Emenity can provide "location based services", which are the detailed localized links and neighborhood information through each hot spot's portal (or splash) screen. Steve Lewis says the idea is to have "a not-for-profit basis for setting up the infrastructure that will later give you business opportunities."
Emenity's biggest customer is the Downtown Alliance, a "business improvement district" (BID) which says its mission is to "create and promote a safe, clean, live-work, totally wired community". There are a number of these throughout the city, and they are supported by a small additional local sales tax. Improvements to Bryant Park -- which are nothing less than spectacular, considering the no-mans-land it used to be -- are an example of a BID at work. Since the alliance servers landowners, it can also approach them with requests to use their roofs or windows for wireless antennas aimed down at public spaces.
The Downtown Alliance's goal is "for nobody to be more than a five minute walk from free local public wireless access". This degree of commitment creates market pull for cheaper and better base-level infrastructure. For example, Terry says, Bryant Park is currently served by a $1000/month T1 line from AT&T, while there is plenty of fiber capacity right next door at the Public Library. So his folks are working to have that bandwidth made available at lower cost and much higher capacity.
When I asked him what bandwidth plans he had for the other locations, he said, "We mostly have DSL lines at the downtown locatations. We're working on a future deployment using the unlicensed 5.8GHz band for point-to-point distribution. The problem is roof rights. It's hard to aquire them, and often expensive -- $6000/month or more. The real estate guys are used to dealing with providers willing to pay that kind of money. But we're making some progress there."
The challenge is always about saving money, time and hassle. And that's the whole idea behind Pebble Linux, whcih Terry created. Pebble is a simple, small, embeddable stripped-down Debian that's easily administered, reliable, and has no moving parts. (For details, see Pebble Linux sidebar.)
Location, location, location
Terry explained why Britt and I failed to serve streets and parks from 28 stories up. Proximity is everything. If you want to set up a public wi-fi hot spot, you have to be right next to the area you want to serve. Line-of-sight is also critical.
At City Hall Park, for example, the rooftop across the street at J&R proved to be the ideal access point location. A square white sector antenna, with a beam width of about 40 degrees, angles down at the park, and provides a signal footprint that serves the park itself and little else. At the far edge of the park, by City Hall, it fades away. A fairly precise footprint also graciously yields to other access points at the local Starbucks, City Hall, the Woolworth Building and elsewhere in the neighborhood.
Terry Schmidt says NYCwireless encourages local citizens operating free access points to label them "NYCwireless" and register with NYCwireless so it appears on the organization's node list.
End user licensing runs the gamut from locked down to free. Time-Warner, for example, aggressively denies users the rights to share bandwidth. At the other extreme, Verizon sells wi-fi access points to its DSL customers. Out in California where I live, Cox High Speed Internet tolerates access points at customer sites, but doesn't promote it. Some local ISPs, such as New York's Bway.net, actively promote wi-fi use by customers. (The EFF maintains a list of wireless-friendly ISPs.)
To the user, the difference between private and public wi-fi service may be negligible. But they are severe differences in kind.
David Sifry, CTO of Sputnik (which makes Linux-based access points), says
The difference between public and private wi-fi service is the difference between a hobby and a profession. And I'm not knocking hobbies. It's mostly a matter of failure tolerance, service level quality, and scale. When you deploy a large number of access points, and they need to be connected reliably, and managed reliably, you're dealing with intolerance for certain gear that's simply handy for an individual putting up one public hot spot in his living room window.
When I asked him for examples of unacceptable components, he focused in on compact flash cards:
Flash is fallible. That fallibility isn't much of a concern in a digital camera, where a few bad bits won't crash a photograph. But a few bad bits in the ROM holding an operating system can crash the whole thing. You need to weigh that risk if you're running hundred or thousands of these things at once."
With that context in mind, it's interesting to see how a company like Verizon roll out large numbers of hot spots, and how that effort puts Linux to use.
The Verizon Horizon
In June 2002, I spent a three days in London taking advantage of free wi-fi served mostly by local citizens sharing their home bandwidth, and listing their nodes with Consume.net, which publishes maps for each, precise to the square meter. One of the most amazing ironies on that trip was the presence of a public phone on the street in front of the access point. The entire weekend I hung out on that street, lots of people used the access point with their laptops, and lots of people used cell phones. Nobody used the phone booth.
Verizon, which has thousands of phone booths on the streets of New York, has seen the same writing on their own wall, and come up with a brilliant plan: Turn phone booths into wi-fi base stations. The first 150 were fired up on May 13, and the company has plans to add the service to 500 or more throughout the city and beyond.
At this writing, the service is available (and free) only to Verizon business and residential DSL customers. But there's nothing in the deployment that prevents the company from opening up to other customers, or completely. It was designed that way. In fact, it was designed to be as easily deployable and modifiable as possible, which is why the company made use of Linux and open source tools. Sean Byrnes, an architect with Verizon, explained it this way:
When we first looked at this, we realized that we needed to manage a lot of devices concurrently. So we wanted a carrier quality solution, which we really weren't seeing. So to us Linux was an invaluable tool to get us running. Normally cost would be a motivating factor. But what motivated us in the early stages was going out, getting the servers we needed, and getting it all set up. What Linux let us do was deploy extremely quickly. So, rather than setting up large servers in one of our data centers, we were able to create Linux clusters and build initial versions that supported the hot spot service extremely quickly, using a wide variety of open source software -- much more quickly than if we had been waiting for licenses, etc.
So we were able to quickly get an entie environment set up, start building the network, building the authentication systems necessary to support it, the management, and the rest of it, before we put it in the data center, which has Sun and Windows systems.
We couldn't have moved it into the data center if Linux didn't allow us to develop with platform independence, and with open source technologies that are implemented across multiple operating systems. We're working to have Linux qualified for the data centers, but it isn't there yet. But starting on Linux is what let us move as fast as we did.
When I said it sounded to me like Verizon was an example of a company that found it easier to roll their own solutions rather than depend on vendors for help, Sean Byrnes replied, "That would be an understatement, actually". He explained,
If you think of very large companies, more often than not, when you're rolling out a new service or application, the argument can be made that the majority of it is glue. Because you already have so many systems and applications already out there that you have to glue them together somehow. So you're forced to be agile. It's never a question of being able to buy a package from a vendor and use it on day one. We had tremendous success using the open source software out there, and finding out ways to use it, change it, and figure out how to use it better, without having to call up a vendor and have their consultants come down next week, stay a couple of days, get under the hood and see what's going on... We could move quickly to build that glue and keep it from bogging us down.
With that many managed access points on the street, the Verizon people have been gaining some valuable experience with wi-fi in the real world. Paul Perry, a director with Verizon, said this:
One thing we've learned from pay phones is that there is traffic. When a bus goes by, it can block a signal. So if you're in a park, you want the pay phone on the same side of the street as the park.
They've also found problems on the client side. Paul Perry again:
We're happy with actual usage. People seem to be able to go through all these paces -- to learn about it, to know what to do on the laptop, to authenticate to a network and finally to get on -- in spite of bugs we have found on the client side. People are getting on with long links and high byte transfers, and we're still not sure why.
Terry Schmidt isn't optimistic about non-free business models for wi-fi. He says, "We don't think that a lot of the for-pay wireless stuff has a sustainable business model. Companies like T-Mobile, with all those Starbucks locations, are hemorraging money, and almost nobody's using them."
Meanwhile, there are plenty of people taking advantage of free wi-fi in places like Bryant Park and Alt.Coffee.
"Free wireless is good for business," Terry says."That's the model. A local business owner says 'I'm going to make my business and my surrounding market more valuable by providing free wireless. It's an attractive thing to do. It enhances the environment and attracts customers.'"
Does Verizon's service, free for existing customers, serve as a conditional flower box? I believe so. Verizon is the incumbent local phone company in New York. It has a lot of home and business DSL customers. Flower boxes that appear magically for those customers are a a nice bonus to existing service. It's a way for Verizon to say Take that laptop out of here. Go sit in a cafe somewhere.
Enlarging civic life
Wi-fi adds a new and practical feature to civic life. For two decades, most personal computing happened indoors, attached to printers, networks, servers and phone lines. If we used our laptops outdoors, it was usually in the same disconnected way we still use them on airplanes.
With public wi-fi, we bring the networked knowledge of the world out into the open air. That changes things.
For all the years I used to visit the New York Public Library, I compeltely ignored the wasteland that was Bryant Park. This last trip was my first exposure to Bryant Park since it was completely re-done in the fashion of the great parks of Europe's cultural capitals. With its lawns, fountains, shaded pavilions and chairs scattered on sidewalks outside restaurants with open doors, it seemed to me the height of civilization. Sitting there with Britt I called it "Paris without the language problem." Bryant Park made me love New York as much as I love Paris, and that's saying a lot. It also made me love civilization and the graces that increase it. That's saying a lot, too.
It is the public places that civilize our cities. Perhaps public wi-fi will civilize the Net as well.
*Wardriving, like warwalking and warchalking, has nothing to do with war. The term is said to have originated withthe movie "War Games" but the less obscure answer is that WAR in this case refers to Wireless Access Reconaissance.