Receive new posts as email.
This site operates as an independent editorial operation. Advertising, sponsorships, and other non-editorial materials represent the opinions and messages of their respective origins, and not of the site operator. Part of the FM Tech advertising network.
Entire site and all contents except otherwise noted © Copyright 2001-2010 by Glenn Fleishman. Some images ©2006 Jupiterimages Corporation. All rights reserved. Please contact us for reprint rights. Linking is, of course, free and encouraged.
Nomad Digital has US trial of its technology: The multi-national firm Nomad Digital is best known so far for equipping the Brighton-to-London run on Southern Rail in the UK with fixed pre-WiMax links. I spoke to Nomad's head the other day, and he said they achieve 6 Mbps in each direction on the local network, but backhaul remains problematic due to limited access to rail right of way. This shouldn't be an issue for Caltrain, the train operator that handles the commuter rail from San Francisco down through San Jose, and on (with limited service) to Gilroy. Paralleling 101, there should be plenty of fiber and other options.
Nomad Digital, Intel, and Sensoria Corp., and Redline Communications worked with Caltrain on the successful trial, which provided continuous service along 16 miles of track between Millbrae and Palo Alto at speeds of up to 79 mph. The test included simultaneous laptop use with streaming video, email, large file downloads, and Web browsing.
Nomad's system is a series of fixed WiMax base stations, which technically means they follow 802.16-2004 and use OFDM rather than mobile WiMax, which is 802.16-2005 and uses OFDMA. This technical difference means that certified equipment is available now. Despite the "fixed" part of fixed WiMax, it's more about predictable placement, as opposed to providing a cloud of coverage as mobile WiMax will offer.
Intel's involvement came from Intel Solution Services, which did something that is obscured by the excessive marketing-speke used in this press release. I think they designed the network, but perhaps they installed the hardware, too. Double-plus confusing.
Update: A savvy reader noted that there is no such thing as certified WiMax in the U.S. market. None of the certified profiles so far include legal U.S. frequencies. Redline's two certified WiMax products use 3.5 GHz profiles, which is still under review here for how it might be shared. (It's available in certain forms in Europe, for instance.) If Redline used 5.8 GHz, that's fine, but that's not WiMax yet and may never achieve a certified profile.
The Bay Area will soon be ringed with Wi-Fi on trains. ACE (Altamount Commuter Express), traveling from Stockton through Fremont down to San Jose, has had service for years, but is finishing a major upgrade that will dramatically increase speed. Their service is free, underwritten by a sponsor. Capitol Corridor, operating from northeast of Sacramento via Fremont to San Jose, has four firms about to launch test networks in expectation of providing a full-blown network across its 170-odd-mile run.
This has got to be the greatest thing since sliced bread for people living in one of the world's most expensive megalopolises. Sure, I'm opposed to sprawl, but I'm also in favor of home ownership, which has pushed those who want or need to work close to the Bay out fairly far with long, long commutes and traffic delays.
Thus, the idea that people might be able to combine public transportation, wireless networking, and reducing their working day by having a working commute seems like the sort of futuristic world--except with no suction tubes moving us from place to place--that sci-fi writers have predicted for decades. Until they went all dystopic on us.
The California rollouts are not unique. Caltrain's trial is part of an explosion of train-based Internet access. SJ has equipped its 42 trains in Sweden; GNER, which runs on Britain's East Coast, will finish equipping its 41 trains this summer; and VIA Rail in Canada earlier this year started their production rollout of service from Toronto to Québec City. [Thanks to Cyrus for the link]
Ruckus announces a gateway design for bridging metro-scale networks into the home: The competition for so-called customer premises equipment (CPE) devices that will bring the weakly penetrating signal of a metro-scale Wi-Fi network into the home is certainly heating up. The once-loved high-gain Senao bridge--not designed purposely for this market--is never mentioned these days, although it was favored. Now it's all about PePLink Surf, which comes in 200 milliwatt (mW) and 400 mW flavors for the home Wi-Fi band, or anywhere from two to 15 times more power than a regular laptop adaptor. (30 mW was once routine, but 100 mW isn't unusual.)
Ruckus will be taking on PePLink and other players for the tens of millions of CPEs that will be sold worldwide as part of the massive rollout of metro-scale networks. Doesn't matter whether you like networks of this scale or not, or prefer WiMax, cellular, or tin cans to outdoor/indoor Wi-Fi. The CPEs will be sold.
The Ruckus MetroFlex Wireless Access Gateway uses its multiple-antenna beamforming technology to pull a better signal out of the air than a traditional antenna design can. The MetroFlex is loud--200 mW--but broad, using six antennas to pull signals out of the air. Four are horizontally polarized and two vertically, which should allow the unit to better handle the extensive multipath reflection that will come with metro-scale deployments.
The firm is claiming -99 to -108 dBm receive sensitivity for 1 Mbps 802.11b and -93 to -96 dBm for 6 Mbps 802.11g. This should allow quite weak signals to be used without sacrificing speed, as metro-scale networks will generally operate at 1 Mbps or slower.
I suspect to avoid prejudicing these networks against older 802.11b adapter that the slowest speeds of 802.11b will be supported--it runs from 1 Mbps to 11 Mbps in four increments--or supported on particular channels. (Metro-scale hardware on and coming to market can cajole different devices onto different overlapping Wi-Fi networks, and speed will be one parameter used to force associations.)
The MetroFlex is designed for remote management, which is going to be a key component, too, of any CPE released at this point. Centralized tools will need to push firmware updates and even reconfigure routers remotely--including, perhaps, forcing a channel change or tweaking power control--to best keep a network adapting to changing conditions, or whenever nodes are added or taken offline.
Right now, the MetroFlex, like other CPEs, requires yet another Wi-Fi node inside the home to redistribute services. There's a 10/100 Mbps Ethernet port to which you would, say, connect a switch, and then connect another gateway or wired PCs.
However, good to note that the unit's specifications talk about multiple SSIDs. More on that when I interview Ruckus for a podcast later this week.
Chandler's Wi-Fi network nearly ready: MobilePro's buildout in Chandler, a city adjacent to its earlier Tempe network, is nearly ready to go live says The Business Journal of Phoenix. Another network in Gilbert isn't far behind that. The three cities combined will cover 187 square miles.
With Vail proceeding, Aspen considers: Aspen is working with Vail to coordinate bid plans for a metro-area network. Vail is working with its telco, CenturyTel, on a deal that will executed soon. That service will feature an hour per session of free service, but new sessions can be created at will. Aspen says they're on a longer timeframe than Vail, and might want a free-only network. With the number of visitors to both cities and across the region, the addition of ubiquitous Wi-Fi would certainly go over well.
Boston suggests that a new or existing nonprofit get the nod to build and own a metro-scale wireless network: This is an interesting idea, which incorporates elements of many different plans. In Philadelphia, the city created (and controls) a non-profit that technically handles the contract with EarthLink and will fulfill digital divide initiatives. Many other cities are signing authorization contracts with wireless providers, which grant non-exclusive access to city facilities, agree that the municipality will assist in greasing wheels of utility issues, and in many cases agree (Philadelphia) or suggest (Portland, Ore.) that the town will shift data/telecom spending from existing contractors to the new quasi-franchisee. (The Boston Globe filed this story; here's the AP story.)
(Update: l spoke on Tuesday, Aug. 1, with Michael Oh of NewburyOpen.Net about the Boston report in this podcast.)
Boston's notion is that by having a non-profit that's not under the city's control, that it will allow many different forms of non-anointed Internet service providers to flourish in selling end-user Internet access at retail. A major underlying goal is ensuring "universal, low-cost broadband access with the maximum possible competition." They would like to see services that cost $35 to $42 per month today cost $15 per month tomorrow without impairing private enterprise, partly by reducing costs to provide such service and expanding the user base. (Update: This story originally assumed that an existing non-profit would be chosen; see the comments from task force member Michael Oh noting that a new, independent non-profit could fit the bill, too.)
The city's three goals are common with other towns: increasing economic development, digital divide reduction, and improving the city's efficiency and quality of service. One metro-scale equipment vendor likes to talk about the pothole problem. If city workers can be equipped with the right form of access, a resident could report a pothole, have that automatically pumped into a GIS system, have the right crew that's already out alerted, and dispatch them to the problem, all in a matter of hours. If you fill the potholes fast, residents think they have good government. It's a starting point at least.
The report presents the stats that, unlike many cities pushing out similar proposals, Boston has nearly 90-percent penetration of the availability of one or more forms of broadband. however, only fewer than 40 percent of households have broadband service, while 30 percent use dial-up. The city already has some interesting digital divide infrastructure: the Boston Digital Bridge Foundation uses corporate partnerships to lets participants pay a computer and printer for $15 per month payments (no interest, no downpayment). And it offers 25 hours of technology training for parents and children.
Whatever nonprofit is chosen or created to carry out this plan will have to raise the $16m to $20m estimated to build and operate the network. The report suggests a combination of donation, equity, and debt. If the city does agree to shift millions in telecom/data spending to the new network, that coupled with, say, an existing nonprofit's endowment and a history of performance on foundation grants should make the money flow from banks and donors.
The city would purchase wholesale access at non-preferred terms, which is a different deal than is in place in Philadelphia, Tempe, and some other cities, in which preferred wholesale pricing and some included mobile accounts are part of the financial arrangements for rights of way and other access. The city will provide infrastructure access such as they own or have rights to, which include parts of 9,000 utility poles owned by Verizon and another utility. This report hopes that this effort would be increased from 800 families per year to 5,000.
The city disclaims cross-subsidization to avoid the political football of perceived extra city services used disproportionately being paid by everyone. Of course, that wouldn't be the case with, say, schools, right? In which school spending has a benefit for the entire city? Naw. But it is a big football.
The nonprofit will not be allowed to offer retail service or customer support, but will be required to run a neutral network. The report emphasizes that national and local ISPs will be granted the same access to operate end-user ISPs. The expectation is that with a 10-percent household uptake--and this doesn't mention anything about business-grade service and so forth--the service could cost $9 per month for wholesale access to 1.5 Mbps downstream and upstream. That's a higher upstream rate than in any typical network currently proposed or under construction. Coverage will be required for all outdoor locations, and exterior walls of every building. This is a hard mark, and I expect we'll see 92 percent or 95 percent tacked on in negotiation.
On the backhaul side, the city has identified that a 50-mile fiber ring comprised of dark fiber could be built out for $2m and connect city buildings in each Boston neighborhood. For user-facing networks, a number of vendors and organizations have also offered to run trials and donate equipment for underserved areas.
The next step is in appointing the city's CIO and the Mayor's Technology Advisor to put together a committee to find the nonprofit partner, and pursue the steps that would allow that partner to build the network.
The Philadelphia Inquirer reports that the city's CIO will exit her position: Dianah Neff has been one of the leading lights in promoting municipally authorized metropolitan-scale networks--note my careful phrasing--and whether or not individuals or companies agree with the concept or her approach, she's been out there swinging with data. Wireless Philadelphia, the non-profit that handshakes with the city on one hand and a private contractor on the other, managed to obtain several significant bids in the earliest process for a city of that scale.
The paper reports some controversy about her administration, including a billing system that cost $18m and doesn't work, but doesn't note other successes. My local county has, at times, spent tens of millions in failed billing or human resource computer upgrades, and it seems to demonstrate more that these complex systems are often hard to manage, and conform to the general performance seen by most IT publications in surveys: big projects spiral out of scale or fail too often. I would have liked to see her department's overall budget, and a small discussion of projects gone right, assuming that she didn't work on two projects for her entire tenure.
The report also indirectly criticizes Neff's frequent travel based on three incidents in her office. But it notes that it's put Phila. on the map worldwide. I don't think I have ever heard so many references to Phila. in the nearly two years since Wireless Philadelphia was conceived as in the previous decades of my life, except, perhaps, to note the Liberty Bell or the Continental Congress. (Man, is the building those men met in small.)
Neff was a good interview, which is one reason for her frequent appearances in the press, because she could talk big picture issues about poverty and training, walk the tech talk on how everything would work together, and run the numbers.
I imagine that good-government rules would prevent her from joining a muni-Fi firm that she had done business with for 12 or 18 months, but there are other fish out there in the sea.
At 50 to 60 rest stops on I95 and elsewhere, pull in for free Wi-Fi: The service won't cost the state a penny, as provide Coach Connect will use ads to support it. Part of the deal allows Coach to put 40-inch plasma screens up showing travel information, and I expect the advertising on those screens will go a lot further than ads inserted into Wi-Fi browsing. It's limited to 30 minutes of free use. The state may install some computers for use by travelers, though goodness knows how they'll secure them.
The article doesn't state how the cut-rate airline will provide cell calls and text messages: It's likely that they are in discussions with OnAir, mentioned late in the article, as that firm is the only one I'm aware of that's prepared to launch in-flight cellular service and, later, broadband over Europe. Ryanair keeps fares low—sometimes as cheap as £1—by charging for every thing, including checked baggage. This allows those who need extra services to subsidize everyone else. Even with a la carte on everything, many passengers pay much less overall.
I thought it would be against the philosophy of Ryanair to install expensive equipment on board planes, such as cellular picocells, but OnAir has been telling me that the retrofitting cost and the weight of their gear are both relatively low. The weight is especially low--about half a passenger's worth. Ryanair's proposed per-minute calls are in range with what OnAir has been suggesting as guidance: £1.80 to £2 (about $3.50 to $4.00) per minute and 40c (70¢) per outgoing text message.
This is a lot of revenue to capture, potentially several hundred pounds per flight, which would easily justify equipment costs. Add broadband on top of that eventually, and there's even more money to be made quite easily. The Irish Examiner makes a rough estimate of £60m for one call per 30m passengers for one minute per flight. More likely, some passengers will talk through entire flight, spending £100 on business calls and other will never use the service.
This article notes that OnAir is now publicly mentioning Air France as an upcoming cell partner--their planned first, in fact. British carrier BMI and Portuguese carrier TAP have both been mentioned before as service partners. [link via TechDirt]
One of many Seattle-based coffee chains switches to free Wi-Fi: The Seattle Post-Intelligencer reports that Tully's will turn 79 stores from fee to free on August 7. Tully's has quite a history with Wi-Fi. They signed up with Cometa a few months before that firm went under, and days away from shutting down their Wi-Fi network, they partnered with NetNearU. That back-end enabling firm didn't operate any locations, and had to reassure their partners that they weren't becoming a hotspot operator.
NetNearU will continue to operate the service, with Tully's paying an undisclosed fee. In many cases, hotspot operators charge $50 per month for managed service with significant discounts for multiple locations. Customer service and technical support calls made by the stores or by the store's customers can be charged on a per-incident basis or on a metered basis to the operator. Service currently costs $3.95 per day or $19.95 per month.
I have heard increasingly frequently in recent months from various parts of the industry (venue, operator, and other) that the cost of charging people for service when it's a relatively nominal amount is often completely outweighed by the cost of billing and the high expectations rendered against the service. There's also a loss of customer when the customer has to choose between free and fee, or among fee-based locations, whether hotel, coffeeshop, or even airport. (Regional airports have started using free Wi-Fi as bait.) With direct revenue not top of mind, it's easier to pay a flat rate and allow customers unlimited usage rather than receive a small trickle of income and have customers have to think about it. It's friction.
Tully's will provide free Wi-Fi in 79 stores in California, Idaho, Oregon, and Washington. They have a handful in Arizona that are apparently not currently offering Wi-Fi nor will do so in the near term. In the Fremont neighborhood around my office in Seattle, we have several independent and chain coffee shops, from an almost literal hole in the wall to a couple fairly giant indies. T-Mobile is about three doors down from Tully's on the main drag (Fremont Ave N.), with a conveyor-belt sushi joint between them.
The LA Times exposes the soft underbelly of metro-scale Wi-Fi (free reg. req.): I've written about utility poles several times before, in that they are the weak link in wireless operators' ability to deploy Wi-Fi or other nodes rapidly. Even if a city or municipality owns the poles or controls the entity that does, they may not operate power to the poles. In this interesting piece in the LA Times, it becomes clear that we're seeing few Southern California municipal Wi-Fi projects because of obstruction by Southern California Edison Company.
Edison has requests a year old for pole access that they haven't acted on. The reporter, James Granelli, notes West Hollywood abandoned Wi-Fi plans, Santa Clarita has been waiting six months for a response, and Cerritos--hailed at one point as a major first unwired city--has only put Wi-Fi in a third of the city's area due to Edison.
In the city of Diamond Bar, officials might use eminent domain to seize one foot of pole access. They're tired of Edison providing no process or timeline. The company claims it needs to test the service, which the reporter notes draw about "two nightlights' worth of power." I might peg that higher--more like the wattage of a reading lamp lightbulb.
In any case, Edison's utility neighbor to the north, Pacific Gas & Electric (PG&E) has no worries. They've set up a flat rate for electrical use and signed contracts left and right. A spokesmen from the company says in the article, "The antennas are low-wattage devices, so even with street lights fully powered, the existing wiring can easily handle the added electric draw. They also have a consistent energy draw, so it's easy to estimate usage."
Edison has suggested that companies might pay cellular mounting rates of $2,000 per month versus the $36/year that operator MetroFi pays elsewhere. Cellular equipment can draw huge amounts of power, in direct contrast to Wi-Fi nodes, and require backhaul connections and other specialized gear. They also have a large revenue profile, and often pay high amounts to building owners and others for specific siting.
The folks at Microsoft's on10 invited me on to talk about Wi-Fi and myself: Very flattering offer. We talked quite a lot of about the frustrations of Wi-Fi and some current developments, and host Laura Foy also touched on the issues of being a journalist blogger. Laura and co-host Tina Wood are both great sports, as you'll see in a segment that was filmed before mine and not yet posted. I can't reveal what's in it; you'll have to check back for a combination of a kitchen product and technology.
on10 is a venture of Microsoft filmed at their spectacular Redmond studios. That sounds like I'm kidding. I've been in television studios before, and on10 and related ventures are quite well set up. The interview had two floor cameras and a remote, as you can tell from the editing.
New York City's parks will eventually be unwired: Dana Spiegel, executive director of veteran community wireless group nycwireless, spoke to me yesterday about the group's efforts to put Wi-Fi in New York parks, the challenges with that, and what's happening in Central Park. We also spoke about an RFP issued by the economic development arm of the city that will examine the state of broadband across all the boroughs and what might be done to improve access to the Internet to all residents. [40 min., 20 MB, MP3]
The city hired an outside firm to calculate whether its investment in applications running over municipal Wi-Fi makes sense in dollars and cents: The firm found that an investment of $26.7m in wirelessly enabled gas and water meters for automated reading would take 14 years to break even and yield $1.6m in savings over 20 years. The city's assistant city manager says the number could turn out to be as high as $8m, and that's before looking at other efficiencies.
The pilot project to test this service found an increase in water and gas usage increased, possibly because of better measurement, but it increased revenue to the city. I might also note that automated meter readings allow businesses to mitigate use at peak times. If gas prices spike or if water availability is low, utilities could subsidize businesses or residents in reducing usage at specific times of day.
The city would like EarthLink to operate their Wi-Fi network. This would save Corpus Christi additional costs in building out the network, and allow the city to receive a revenue share as well. Correction (see comments, too): EarthLink is only talking about a pilot program to provide retail service to residential and business customers.
Slim Devices will ship the Transporter next month for audiophiles: the $2,000 streaming media device is designed to handle uncompressed and lossless music formats with aplomb, supporting several forms of digital input and output to integrate with high-end stereo systems. Should I just say it again? High-end. High end, high end, high end. Anyway, if you can hear a pin drop at 1,000 meters, and hear the missing notes on CD recordings, you might want--nay, need--the Transporter. There's a lot of detail on their digital-to-analog converter and the accuracy of their signal timing for those who understand such things. If you do, you're a candidate to buy it.
Along with handling optical, coax, BNC, and XLR digital input and output, and playing back files stored on locally network computers in Apple Lossless, FLAC, and WMA Lossless--along with every uncompressed and compressed format out there--the Transporter includes auto-switching/auto-sensing 10/100 Mbps Ethernet, and 802.11g with dual external antennas. They offer WEP and WPA/WPA2 support (personal flavors).
Boeing released its second quarter earnings, and stated Connexion's drag: The company had a loss of $160m largely because of one-time charges, including a $615m payment (after reserves, a charge of $571m) to the U.S. government. Boeing said that if it were to sell or shutdown Connexion, it would affect its earnings by up to $350m before taxes. The company also stated, however, that it would see an increase in earnings per share of $0.15 in 2007 from Connexion being off its books, which translates to roughly $120m per year. (The earnings-per-share figure is in the conference call, not the press release.)
Did they just back out the operating expenses net minimal revenue of Connexion? Or is that too easy? Some analysis put Connexion's annual costs at $150m to $400m two to three years ago. They may have pared costs by shedding expensive transponder leases and by cutting expenses through efficiency, too.
During Boeing's earnings conference call, a Wachovia anayst, Joseph Van Pietro, asked a great question. Connexion has been billed as an operational advantage for Boeing's aircrafts, especially the 787. With Boeing no longer owning or having Connexion available, how does that affect future operating costs for new aircraft?
Boeing's chairman and CEO, Jim McNerney, bluntly stated that "our business model is not being met," and that "continuing to operate as we are now is not an option." Restructure, terminate, and affiliate are the three "obvious options," he noted. "All elements of the business model are falling short of the projections."
Pietro also asks about how airline customers are reacting, and the CEO punts on that front, but puts some of the blame on airlines who have not adopted Connexion. This has been the problem ever since the U.S. domestic airline market tanked post-9/11. Connexion model had thousands of planes equipped and billions in annual revenue. Instead, they have hundreds and millions.
Mobile WiMax explained for you: In this podcast, I interview Monica Paolini of Senza Fili Consulting. Paolini is an expert on the intersection of financial projections, technology, and market needs, and works with the WiMax Forum among other groups. Because of her wide-ranging interests, we focused on mobile WiMax in this podcast, as I think it's one of the coming technologies that's worst understood and most misrepresented because of the obscurity of parts of its operation.
We talk about the differences between fixed and mobile WiMax, along with 802.16-2004 and -2005, the underlying standards that are commonly associated with the two terms. We walk through the spectrum bands that might be used in the US and internationally for both fixed-only and fixed/portable/mobile services. And Clearwire's recent massive receipt of Intel and Motorola money is examined as it affects the future of mobile WiMax in the U.S. [39 min., 20 MB, MP3]
Oono Transmita offers generic wireless audio link: The paired receiver and transmitter will pass audio from any stereo output jack, such as that used for headphones, to a like input jack--say on a stereo receiver with great speakers. While there are plenty of audio devices on the market that stream digital content from a computer or via Internet radio to a stereo system or through built-in speakers, this is a truly generic audio bridge. It uses 2.4 GHz, but isn't Wi-Fi. The transmitter can be powered via AC or USB (though no data passes); the receiver requires AC.
The paired devices is sold for £34.95 via a UK firm that specializes in digital audio broadcasting (DAB), the terrestrial digital radio standard widely used in the UK and barely used throughout the Continent. (In the US, an incompatible technology known as HD Radio is just making inroads. I've started an HD Radio blog to document those inroads.) The item doesn't appear to be available to ship to the US from this store because of that DAB focus.
The Boingo aggregated hotspot network encompasses whole city: It's the beginning of things to come, in which entire metropolitan-scale networks are added to the roaming networks of hotspot aggregators, or are part of peering arrangements among like partners, city to city. Taipei's Wifly network, which spans most of the town with Wi-Fi, will be available through a single account using Boingo's software. This isn't free roaming; most of Boingo's roaming outside of the U.S. involves prenegotiated, discounted fees. All Boingo's locations in Taipei, for instance, are currently tagged as Premium Locations with per-minute charges.
The popular press keeps dubbing WiMax "Wi-Fi on steroids": Ofcom, the British communications regulator, is considering turning Wi-Fi itself into Wi-Fi on steroids by allowing a 100-fold increase in direct power for the 2.4 GHz and 5 GHz bands. In the U.S., the limit is 1W (measured as a calculation of the radio, amplifier, signal loss, and antenna) and 4W effective output for 2.4 GHz, and a variety of rules for 5 GHz. Ofcom is considering 10W, while the European Union might allow 4W. There might be some minimal licensing requirements for the higher power plus a requirement of presence to avoid using higher power where it's not allowed.
This article points to a European technology blogger who noted the Ofcom plan, and who writes about the higher limits, that the plan "completely ignores the health risks as 2.4GHz is the frequency microwave ovens use and 10W though low power could have significant health risks." Let's just call that bloody ignorance, shall we?Microwave ovens reflect a signal in a confined space to change the magnetic polarity 2.4 billion times a second of water molecules; the twisting thus causes friction and heat. Microwave ovens typically use 800W to 1200W of power. Beaming out microwaves doesn't produce an oven effect because it doesn't have the reflection.
However, there are extremely well known health effects from close proximity to high-power microwaves, and 10 watts counts. Even in the U.S., you can produce signals of nearly 30W under certain directional rules using certain kinds of antennas. It is never clever to position oneself closely to a high-power antenna--something at least an order of magnitude higher in output than a home gateway. High-powered Wi-Fi or other wireless uses directional antennas typically (because the rules don't allow high-power omnidirectional use). These antennas are typically on roofs and poles or pointing far away from human beings.
The firm has Toshiba and Sony among initial advertisers, MetroFi among initial operators: JiWire, with which I have a long-standing relationship, has extended its advertising sales from its own highly trafficked Web site--which features a worldwide hotspot directory and articles on using Wi-Fi and cell technology--to hotspot operators. JiWire has announced that nine operators will be part of the first rollout of their advertising service, although they only list MetroFi.
MetroFi is one of the three significant national metro-scale wireless infrastructure builders, the other two being MobilePro and EarthLink, judging by contracts signed or in process. Only MetroFi maintains that they can run networks profitably by offering free, advertising-supported accounts. EarthLink and MobilePro have specifically dismissed that possibility.
I tried to find other active hotspot-oriented advertising networks and came up empty. While there are firms that offer the technology to make hotspot advertising work--such as PerfTech, which counts ad serving technology (but not sales) among a variety of services they offer, and the equally diversified Front Porch--other companies handle the ad sales. And there are vertical-industry targeted ad networks that handle, for instance, car dealerships that have hotspots.
But JiWire's entry might provide validation for some of the model, not because JiWire is the be-all, end-all, but rather because JiWire has a relationship of some kind of virtually every hotspot operator in the world, and has three years' experience in selling advertisers and ad agencies on targeted ads on Wi-Fi topics. We'll see how it plays out.
I'm particularly interested in getting JiWire's rate card, so that we could know what rates they're charging--which could help prove or disprove MetroFi's long-term plan, too.
[Disclosure: I hold shares and options in JiWire, and have an ongoing editorial relationship with them. I have no obligation to write about the firm, nor to write positively about them.]
When you're swimming in the ocean, even the lowliest tissue becomes a water detector: Lo, a few years ago, when we were lonely islands of Wi-Fi connectivity in an uncaring void of wired broadband, cruelly locked away within homes and businesses, a Wi-Fi detector acted like a beacon, a lighthouse if you will, giving us some hope for a seamless tomorrow. Early Wi-Fi detectors lit up in some fashion, often using multiple lights to indicate a range of signal strength, when Wi-Fi was near. Later detectors added directionality, and then LCDs to show network names.
Now, drowning in the Wi-Fi surf, as it chops about our head and shoulders, the undertow sucking us deep, deep, deep linked down, full fathom 101--that's binary, folks--thy father lies, of his cables are coral made, those are radios that were his eyes, and so on, a Wi-Fi detector doesn't tell us more about the deeps. A solid, unblinking light, like the all-seeing eye of Sauron (cf., Lord of the Rings) just lets us know we're within the scope of Wi-Fi, not whether it's open, free, available, or, as is more likely, shut to the likes of us.
The pen is mightier than the sword, but you can't use a sword to determine whether a Wi-Fi network is around. Unless you threaten the nearest geek with it, I suppose. For $18.95, a nice, limited edition paperweight which, in most areas, confirms what we already know. Such as, my window is a rain detector--I open it up, lean out, and if I get wet, it's raining. [link via BoingBoing]
Microsoft confirmed it will offer an iPod competitor: The player will be equipped with Wi-Fi, ship later this year, and be called Zune. (Inside joke for moviegoers this weekend: At least they didn't call it a Narf, which is only marginally less euphonious.) Some content will be sharable among devices, which sounds a bit like the Music Gremlin. Microsoft will be making this device.
A Champaign-Urbana-based group gets $500K from the National Science Foundation: The Champaign-Urbana Community Wireless Network (CuWIN) group in conjunction with the University of Urbana-Champaign (UIUC) will use the grant to continue to develop their open-source mesh routing protocol system over the three-year life of the grant. The group wants to produce software that creates self-configuring, self-healing mesh networks with advanced properties that would allow clusters of nodes in which some cannot hear all others can still function with a high degree of reliability.
Toronto Hydro owns the utility poles, even, and they're still a problem: Toronto Hydro bought the street light system from the City of Toronto, ITBusiness.ca reports, and they were unaware that their plan to bring Wi-Fi to the financial district would be snagged by problems at the poles. While they had location candidates for placing Wi-Fi nodes, they found that many poles in the financial district lacked power at all hours of the day; this is often the case for street lighting. They had to upgrades those poles. In other areas, they found an absence of poles entirely, which required working around a lack of assets.
Law enforcement also raised objections to the telecom arm of Toronto Hydro's plan to offer free service in the first six months of operation. They wanted user tracking to avoid "illegal or unsavory" network uses. Instead of applying a direct tracking, they came up with a compromise: You must supply a cell phone number to which the system uses SMS to deliver a user login and password. (For international travelers, that means "free" service carries a US$1.00 to $2.50 surcharge, approximately, for international SMS.)
When the system goes to a for-fee basis, they'll use a different authentication method that will still require some kind of physical personal identity confirmation.
The specter of electromagnetic emissions also raised its head, but Toronto Public Health gave the utility a green light.
The goal reamins to cover 600 square kilometers over three years.
Does this mean 1 Mbps downstream for free is unsustainable? News.com reports that MetroFi, one of the three leading U.S. metro-scale Wi-Fi operators, has partnered with AT&T for a bid on the 65-square-mile Riverside, Calif., network. In what might be a footnote or might be a significant development, this bid offers only 512 Kbps downstream at no cost, unlike all of MetroFi's other bids and operations, which offer 1 Mbps downstream. That rate is available from AT&T at $20 per month, News.com reports. (Upstream rates are a mere 256 Kbps in both cases, unlike most of the paid services in which 1 Mbps symmetrical is offered.)
News.com quotes MetroFi's head as noting that they need the "brand recognition and the financial resources of a larger company like AT&T, which we could use to market the service." This makes little sense to me--if you win a contract as what is essentially the exclusive Wi-Fi operator for an entire city, along with which comes in some cases various advertising rights, a citywide network signal that anyone with a laptop or handheld can see, and the local press coverage that these networks are garnering--why is marketing an issue? Further, Haas told me months ago that eliminating marketing expense was one of the reasons to go free, as it reduced costs.
This has received wide play for some reason: The city council gave their official okay for the estimated $10m contract, the cost of which will be borne entirely by MetroFi. That firm said that they will start construction of the network immediately, building it over a period of two years. One losing bidder, VeriLAN, expresses skepticism over the free, ad-supported service model. And Personal Telco's head expect interference to be worse than advertised. The service is 1 Mbps down and 256 Kbps for free with a required ad banner at all times. A $20/month subscription removes the ad banner. Higher speeds should be available for some higher cost, although that wasn't discussed in this article.
One issue about ad-supported networks that hasn't been widely discussed is that the revenue model usually looks at substituting full-time, heavy monthly users who would otherwise pay $20 retail to a firm like EarthLink, MobilePro, or MetroFi getting service for free. But this ignores what will likely be a very large audience of occasional users who wouldn't pay the day rate for the service, and who might not like what is typically a ratcheted down free service in cities that also have for-fee services--they're usually slower, time limited, or have other restrictions.
Thus there's a residual additive ad revenue stream by capturing eyeballs from visitors, suburban residents who come in occasional, or those who have a high-speed connection at home and are only mobile at times. That may be part of the secret of cracking ad-supported free service, if that nut is profitable enough to crack in the long run.
Google maps its Mountain View coverage: The map shows where all the APs are and where coverage holes have been identified. Recall that Mountain View isn't designed for indoor coverage. [link via GigaOm via Niall]
Inc. magazine writes about entrepreneurship, and they give Tropos kudos in this article: Most revealing is revenue: $15m in 2005, $45m projected for 2006. With nodes costing $3,500 (retail, mind you), the numbers noted by EarthLink and others for Tropos-based deployment are pretty easy to use to project additional revenue. What EarthLink pays, of course, isn't retail, when they're buying 3,000 at a go. It also means that EarthLink will conceivably form an enormous portion of their current revenue, and relying on a single customer is always problematic for businesses, which try to diversify suppliers and customers.
The story as a whole is really about Narasimha Chari and Devabhaktuni "Sri" Srikrishna, the company's founding engineers, and the guys behind Tropos's mesh protocols. It's a charming profile of how they went from an idea into garage-scale implementation, and then into funding the company. (Tropos was founded as FHP Wireless, and focused initially on enterprise-scale campuses; I talked to them about four years ago when I was researching tools for managing large numbers of APs at corporations.) The narrative picks up on CEO Ron Sege's entrance into the company about halfway through.
There are a few telecom and technical problems with this business article that do affect the business case that's made. The writer notes in regards to what would keep Wi-Fi from working outside that there's no bar: "Install 30 of them [nodes] per square mile (which isn't hard, since an installer using a single tool can put up a unit in 15 minutes) and they immediately begin communicating with one another via radio waves." Of course, 15 minutes doesn't include the permitting time to obtain pole access, the RF simulation to sort out where nodes should go, nor the time nor expense in putting power at the pole, not always an easy situation.
"In the United States, most towns already own the infrastructure for suspending 14-pound boxes in the sky: lampposts, traffic lights, telephone poles, city buildings." That's really not true. Utility poles are the main hanger for most metro-scale nodes, and only some cities own some poles. There are places in which a city runs all the utilities. But from my reading, most towns and cities have more complicated stories, including a mix of two or more public and private entities that control the pole rights. While the Telecom Act of 1996 requires non-discriminatory access to poles at reasonable prices, the parameters are broad enough that pole access can be a gating item for deploying any kind of telecom or cable service.
"The Wi-Fi cards that early adopters were sliding into their laptops in 1999 went for about $2,000 apiece.": Apple and Lucent were selling cards for about $100, and even enterprise cards were under $300 in 1999.
While the account of the Pennsylvania law that almost scotched the Wireless Philadelphia network is mostly accurate, the conclusion is not: "In late November 2004, just as the bill was approved, Philly's Wi-Fi enthusiasts got a break. 'It was almost like diving to get the catch in the end zone,' says Sege. The state agreed to exempt Philadelphia from the requirements. (All other Pennsylvania municipalities remain bound by it.)" Actually, Gov. Rendell was able to get Verizon to sign a waiver that conformed to the terms of the bill before the bill was signed; he then signed the bill. And it turns out that other municipalities in Pennsylvania are apparently able to have private firms build them metro-scale networks; they just can't do it themselves without petitioning the incumbent telecom.
The outcome, however, the reporter gets exactly right: "The way Sege sees it, Verizon's in-your-face tactics were the best thing that had ever happened to the start-up. The giant telecom's reaction made dozens of other cities take notice. If Verizon was so ruffled, people seemed to think, then Philadelphia must have been on to something interesting; the technology's potential must be real." There's a nice bit that follows about how EarthLink was recruited into the muni Wi-Fi business, something I haven't heard before.
The writer doesn't know who pays for this networks, unfortunately, nor that customers aren't going to come from incumbent telcos and cable firms: "After a city government invests $20 million, no users will be happy if their connections go down or their webpages load slowly. The last thing Tropos needs is for annoyed customers to head back to Verizon." In Phila., EarthLink is investing the cash.
Overall, a nice article, only slightly on the puffy side, emphasizing entrepreneurialism without discounting the technical challenges, covered in depth at the end.
BSkyB is latest to offer free broadband in the home: Free needs an asterisk, as the Sky Broadband service requires a subscription to the satellite television service (about £17/$31 per month), a £40 ($73.50) activation fee, and may involve some installation fees. The service is ADSL with 2 Mbps downstream, and will include a free wireless router. A 16 Mbps downstream service is just £10 ($18) per month. And a landline VoIP-like service will add £5 ($9) per month for unlimited calls to landline numbers in the UK. BSkyB will put £400m ($735m) into its broadband service over the next three years, expecting profitability in 2010. (A reader notes that this isn't VoIP per se: It's Carrier Pre Select (CPS), in which wireline calls are terminated via Sky's network. This is similar, I believe, to US providers like Speakeasy which use VoIP as a carrying mechanism over the local loop.)
BSkyB isn't alone, which sounds odd to those of us outside the UK. Television service and mobile phone offerings are in such a high state of competition, that broadband-included packages are not unusual. Carphone Warehouse offers 8 Mbps ADSL with a £20 ($38) per month unlimited UK landline voice package (that's the £11 line fee plus their service), a £30 ($54) activation fee, and an 18-month contract. An extra £1 per month gets you unlimited landline calling in 28 countries. (The British advertising watchdog ruled that Carphone can't call this service "free" because the broadband service is integral to the voice offering, but Carphone has finessed this and can advertise free by adding the broadband to previous calling plans at no charge.)
Orange offers the same 8Mbps speed with a £30/$55-per-month, 18-month contract, and has no activation fee.
The rail operator on Britain's east coast has accelerated its plans to put Wi-Fi on board: GNER will have Internet access from cellular and satellite distributed via Wi-Fi on all 41 trains in its fleet. They've completed installing service in their 30 electric Mallard trains, and are finishing up by adding access to the remaining 11 trains, all diesel. The original plan was to deploy by May 2007; now, it's by the end of summer due to customer interest. First-class passengers get free access; coach pays.
UK hotspot builder The Cloud has opened the first of several city center hotzones: The firm is building city center networks using Wi-Fi across nearly 10 cities, most of which are partly operational, ZDNet UK writes. The City of London (the square mile business district) is one of the most notable of these hotzones. Service runs £11.99 for a week or, with a year's commitment, £11.99 per month. The Cloud resells its network to many aggregators, including BT OpenZone and iPass, resells VoIP access to Vonage, and offers use to Nintendo DS game player owners.
Wireless Week reports that the low-power, long-lived ZigBee standard is gaining momentum: ZigBee (802.15.4) is designed to provide small amounts of information over a very long period so that it can be embedded as radio technology in sensors and remote controls. The idea is that devices that otherwise would require wires in order to produce readings could deliver telemetry or send very small control strings and not require battery changes for many months or even years.
If ZigBee became widespread, it would dramatically reduce the need for wiring for heating/cooling in hotels and office buildings (and eventually in homes). It would also allow many incompatible systems to be replaced with one simple standard that will be cost effective to manufacture in bulk.
Right now, ZigBee single chip chipsets are $3 to $5. It needs to get below a buck. Ethernet inventor Bob Metcalfe, the chairman of a key chipmaker, noted that 10b microcontrollers are sold every year, and ZigBee could be significant part of that market.
I'm trying to make sense of Ruckus's rural strategy for its IPTV products: The company uses multiple-antenna technology combined with proprietary streaming algorithms to provide voice, video, and data (802.11b/g compatible) across a home. The rural angle is intriguing, because rural telephone companies want to bring newer services (and higher per-customer revenue for the same wired infrastructure), but they can't afford to rewire homes to handle the network for multimedia and VoIP traversing a house.
Enter Ruckus. They say that because their system can carry streaming video and deliver other services, they're the perfect complement for rural telcos. The telco still does a truck roll, but Ruckus claims its MediaFlex system of gateways and adapters takes under an hour to install, and future additions can avoid a truck roll.
An hour is a pretty nice bar to set to keep costs low, and compares favorably to other home installs. A DirecTV installation at my house required two installers and about an hour to mount a satellite antenna, set up the receiver, and train us on the system. Obviously, the satellite industry considers an hour a profitable installation when factoring in lifetime customer value.
By contrast, in a DSL textbook I read nearly a decade ago, new telco services weren't considered profitable by large phone companies until they reached the point when only five percent required truck rolls. It took DSL and cable years to reach the point where most installs involve just sending a modem out. This has changed completely again with triple-play services, as Ruckus notes.
The latest press release from Ruckus notes 16 more rural telcos in addition to several they'd already signed.
Wayport has built Wi-Fi into 8,000 McDonald's, and I talk to Wayport CEO Dave Vucina about that in this podcast [38 min., 19 MB, MP3]: The hotspot infrastructure company has reached a goal set two years ago at the introduction of its Wi-Fi World plan. Typically, hotspot operators resell access to their networks for a per-session or per-user fee that aggregators and resellers pay them. Wi-Fi World proposed a single, fixed monthly rate for an entire network (on a per-location basis) that would be invariant based on use.
AT&T (then SBC) was the first firm to sign up, and acts as a broadband and services provider for Wayport, too. Nintendo followed this last spring as a way to offer free Wi-Fi access to players of its DS system. With a per-session fee, costs get higher with more usage. With a monthly per-location fee, the more users, the cheaper per-user cost.
Vucina and I talk about a range of hotspot operation issues, including how Boingo's acquisition of airport wireless infrastructure builder Concourse Communications affects Wayport's model, how Wi-Fi World has played out, the growth of the applications business, and how Wayport might get into the locally cached content business. It turns out, it's all about applications, as in many other maturing wireless data businesses.
Nortel's Richard Lowe is operating from some strange premises: In this News.com editorial column, he states a variety of unsupported or simply odd conclusions. The article is primarily an attempt to defend cell operators against Google, for whatever reason I can't suppose. There must be some conflict there. The stalking man here is free Wi-Fi. While most municipal projects will not offer much or even any free Wi-Fi, that's the subject Lowe has fixated on.
He suggests that Google and EarthLink are collaborating on the San Francisco municipal Wi-Fi network, which is partly right (they submitted a proposal, but EarthLink will run the network and Google buy service to offer for free). But he notes Google will probably run ads, which the company recently said they would not in the early phases of its Mountain View network, or perhaps ever; this might translate into their SF plan, too. He talks about "combing through ads," which implies he's never used Google, and implies that Google would use interstitials or other invasive mechanisms.
He says that free service would be a lower quality and "less secure," which implies that Google and EarthLink can't operate a secure Wi-Fi network. In fact, Google is planning to offer a free VPN, and EarthLInk has stated several times that they'll use a secure local link (via EAP-TTLS, a secure Wi-Fi/wired authentication technology). The Tropos nodes on EarthLink's networks will communicate via a secured protocol, too, from what I've been told.
Lowe then ventures into an area he clearly hasn't studied or would like to represent outside of what we agree is reality. "Wi-Fi is great for enterprises and municipal hot spots...But Wi-Fi signals travel only several hundred feet. So unless you have hot spots linked across your city, you will not get a constant connection as you travel. You will also have to sign on each time you change hot spots." Woof. I guess he doesn't know what a metro-scale network is: a large, enterprise-like network. He is, instead, writing about free hotspots that might be offered by a city or downtown commerce group.
He notes, "And unlike 3G cellular, Wi-Fi is not built for full mobility." This is true if you measure mobility as "traveling fast." There's fixed service, in which one is always at the same locations; portable or nomadic service, in which you carry an adapter to different locations; and mobile service, often defined as vehicular traffic, as on the FCC's 3G Web site. (The FCC says 144 Kbps in vehicular motion is still 3G! This might be a bit out of date.) Wi-Fi can provide fixed and nomadic service, and mobile service often works up to a decent rate of speed, but requires some cleverness to work at highway speeds. That's a big area of development, and it's one where the cell companies have Wi-Fi beat solid.
This is where we get weird. Lowe defines why 3G is great versus Wi-Fi. "It offers ubiquitous coverage, traditional phone services and advanced Internet Protocol services like instant messaging, picture sharing, mobile video and interactive games. And it does so at greater distances, more securely and with higher quality. Unlike free Wi-Fi, there is no limit to whom or where you can call, e-mail or otherwise communicate with in the cellular world."
I can't argue with ubiquitous coverage, but it's getting harder to define what traditional phone services really means when VoIP is offered by so many major carriers worldwide. The advanced IP services? Don't get me started. Cell operators have walled gardens. When using Wi-Fi, you have the entire Internet at your disposal. The notion that cell operators provide unfettered calling (can't call VoIP numbers via SIP from a cell phone or use Skype) is a laugh, too. And your cell phone can e-mail anyone where free Wi-Fi doesn't allow you to? I don't get that point at all.
Now we get into a misunderstanding of bandwidth. "Wi-Fi simply cannot accommodate growing consumer demand for ubiquitous, immediate device-agnostic content and services delivered in the most simple, entertaining and reliable way." Wi-Fi isn't a delivery mechanism in the way a cell platform is. When he writes that "Wi-Fi" can't accommodate this, he might mean free Wi-Fi, but even so, that implies that spectrum-constrained and service-agreement constrained 3G services will deliver better results than, say, a cheap DSL connection in a cafe. Hard to see that. No YouTube allowed on Verizon, for just one instance.
Lowe doesn't ignore telephony. "The current Google Wi-Fi offer has other limitations as well. For one thing, the "free" phone service primarily works for PC-to-PC calls. Call your friend's cell phone or BlackBerry from your Google Wi-Fi connection, and it will cost you." Right. Again, we're focused on free. And my "free" cell phone service that I pay $100 per month for (for my wife and I) includes 850 minutes of prime time calling. For $100 per month, I could get 5,000 minutes of 2-cent-per-minute calling. So it's a hard comparison.
"Another thing to keep in mind is that most PCs were not designed for phone calls. Most PC and laptop speakers, as well as microphones, do not provide the level of voice quality we've come to expect from our phone service. Moreover, laptops aren't the most convenient devices to carry around for making phone calls and connecting to the Web on the go." Oh my goodness--would someone tell this man about Treos and smartphones, and how every single handset maker on the planet has or will shortly have dual mode cell/Wi-Fi handset? Stat!
"IMS [a cell network multimedia standard] transforms the Internet from a static document storage and retrieval tool to a more interactive, entertaining and "live" environment with real-time services." Mr. Lowe, by the way, is running Windows 95, otherwise this statement comes from Bizarro world. "Me like multimedia, so me not use Internet to access it."
This man needs our help, not least of which because Nortel builds Wi-Fi and WiMax infrastructure that directly contradicts his entire column.
Dell will offer Broadcom-based Draft N adapter as built-to-order notebook option: The "Dell Wireless 1500 draft-802.11n dual-band wireless card" will use the Intesi-fi technology that Broadcom has developed in advance of an industry-approved standard for 802.11n. Broadcom isn't alone, but I'm stunned that Dell will sign onto this at this stage. The upgrade costs $59. (Acer will ship a Q3 laptop with Draft N built in, The Register reports.)
The press release from Broadcom states, "Broadcom Intensi-fi technology complies with the current IEEE 802.11n draft specification and is available in a variety of draft-802.11n routers, including those from Linksys, NETGEAR and Buffalo." There is no way to comply with a draft specification of this sort. It's an early draft, likely to change, and there's no one outside of the firms trying to push this early Draft N gear who believes it's a good idea to write one's name in water.
The Broadcom press release also states, "Intensi-fi solutions are also interoperable with draft-802.11n technology from other chipmakers." Yeah, right. In certain testing which belies most of the magazine lab tests of the technology. What's the brand promise behind this statement? What happens if a competitors updates their firmware, and interoperability fails? This is why the Wi-Fi mark works--stable standards, independent lab testing, and the possibility of failing tests--and this kind of standards-by-marketing committee fails.
This is making me slightly ill as I see companies rush to push something out that nobody needs. Regular MIMO on the market provides the distance boost that's really at the crux. The rest of this Draft N technology could patiently wait until the standard is done.
I reiterate that no manufacturer I'm aware of is willing to promise that equipment they release today will be fully upgradable and interoperable with the final, release 802.11n specification even if they have to swap out hardware. Without that promise in place, they're selling what could turn into expensive paperweights that offer minimal functional improvements at excessive cost compared to what final, shipping, interoperable, certified products will provide in probably no more than six months.
Wait, I say, wait.
A Dell spokesperson provided a clear statement that I believe is frank and fair to my question as to whether Dell would offer upgrades if hardware were required. Dell said,
"Dell felt there was compelling value for our customers in the current draft standard, in terms of range and throughput, to justify releasing a product based on the draft.
"Although the Dell Wireless 1500 is fully compliant to the current draft and several elements of the draft will be incorporated into the final standard, Dell cannot guarantee upgradeability to the final standard. Regardless of final upgradeability, the Dell Wireless 1500 card will continue to perform at throughput rates and ranges superior to 802.11g, when paired with Draft 802.11n routers with the Intensi-fi technology, and provide customers with the ability for multiple users to use high-bandwidth wireless applications throughout the home.
"Also note, the Dell Wireless 1500 Draft 802.11n card is backwards compatible with 802.11 a/b/g wireless standards, so users will always be able to access these wireless networks no and in the future."
This is well stated. There is nothing misleading or incorrect in this response. However, I don't believe that any Dell customer should purchase what is essentially a beta or pre-release item that cannot be guaranteed upgradability. But I appreciate that Dell isn't overhyping the product.
You can always read more about 802.11n and MIMO and MIMO+N Networking News.
PC World reports that a few hundred Trusted Testers will start providing feedback on Google's Wi-Fi network: The network won't carry advertising, will require a login, and isn't designed for interior use--although a bridge might allow that. Google is paying all the associated costs. They'll release their VPN client, at long last, to use over the network (Windows only at first). PC World says that Mountain View residents have received a "blizzard of promotion deals" for broadband service from Comcast and AT&T.
Tropos releases a software update for its mesh-networking hardware designed to improve interaction with handheld devices: The company said in a briefing earlier this week that Adaptive Mesh Connectivity Engine (AMCE) tweaks the mesh nodes' approach to dealing with different end-user devices on a packet-by-packet basis without any changes in the client hardware or software. Each packet, Tropos said, can have a different power output level, and can be customized with specific timing and framing characteristics that work best for a given device. The software is available today to Tropos users with support contracts. No hardware changes are needed.
Ellen Kirk, Tropos's vice president of marketing, said quite accurately, "There is no such thing as a standard Wi-Fi client." (It's also true that not all 802.11a/b/g devices are certified Wi-Fi, and thus a "Wi-Fi" device might not be that at all--it might not conform to interoperability standards required by that mark, which makes a "standard Wi-Fi client" an even harder mark to hit.)
The software also mitigates interference by analyzing the radio frequency noise floor and working around that to better "hear" incoming traffic and produce a usable outgoing signal of the right strength.
Later this year, Tropos will release an update that will use another bag of tricks--some of which I have heard of being deployed in enterprise-scale wireless LAN switched networks--that can force clients to associate with particular nodes rather than allowing the client its choice. While this is a hard-wired MAC (Media Access Control) issue that is very dependent on a particular implementation of 802.11 on the client, there are ways that access points can be clever enough to fool the adapter and make it stay put instead of hopping among multiple choices.
Earlier this year, Tropos released a set of extensions (TMCX) for customer premises equipment (CPE) device makers that would allow the CPE to talk to nodes and negotiate parameters and pass reporting and provisioning information. AMCE doesn't require CPE coordination to produce benefits, Tropos executives said, but rather reduces client variability on the network side by adapting to the Wi-Fi client's needs.
George Cooper is a patient man: He became CEO of OnAir, a company formed by shareholders of Tenzing alongside Airbus and airplane systems integrator SITA, in Feb. 2005. At the time, the expectation was that the satellites that their new in-flight cell phone and broadband service would rely on would be launched by Inmarsat by early 2006. The current reality is two birds in the air as of several months ago, but the third satellite--covering the critical Trans-Pacific routes--won't loft until 2007.
Ah, patience. I talked at length with Cooper about the origins of OnAir, the issues of providing satellite-based service, how fuel prices have affected the business of adding equipment to planes, and the upcoming service additions. Fuel turns out to be critical. The GSM/GPRS component that OnAir can provide weighs 42 kilograms. Their full Internet access system has a variable weight dependent on what's on a plane and other factors.
Right now, OnAir expects to offer GSM and GPRS service for multiple airlines starting in the first quarter of 2007, with more coming online as the year progresses. The company is also dependent on the Airbus A380 production schedule, now in flux after delays, as that plane will include the "Live IP" or Internet access equipment as a prewired component that airlines can opt for. Pricing for GSM/GPRS will cost about that for international roaming, or about $2.50 a minute. However, the price will drop 10 percent per year over a five-year contract per their agreements with airlines.
I joked that at these rates, a single call made on, say, Ryanair (not an announced partner) could cost more than the fares paid by all passengers on the plane. Cooper noted that the market has changed with low-cost carriers who obtain their revenue from sources other than strictly airfares. [36 min., 17 MB, MP3]
Wired News reports that Brian Salcedo will stay in jail until at least May 2011: The convicted network intruder received a nine-year sentence for his part in using a Wi-Fi connection to break in to home improvement chain Lowe's computer system and alter software in an attempt to capture credit card numbers. A partner received 26 months, and has two months left in his sentence; he's already in a halfway house.
Salcedo was sentence not based on success--he never even saw any of the six credit card numbers captured during the brief window in which their software was running before Lowe's disabled it. Rather, he was given a sentence based on outcome, if their scheme had worked. Wired News reports that an appeals court uphold the judge, noting the lower court didn't err in using intentions and potential to increase the sentence's length.
SkyPilot has raised another $21m to continue growth: The company is an early entrant into metro-scale networking, once lumped together as mesh networking, but now involved a variety of approaches of which SkyPilot's is currently unique but analogous to methods employed by BelAir, Cisco, Motorola, and Strix. Their equipment is used by MetroFi, which has been scoring contracts left and right, and thus one might see a connection between the upcoming demand for millions of dollars of equipment by MetroFi and SkyPilot's ability to raise funds. However, the company talks about a large number of customers--15,000 units to 200 customers in 40 countries--MetroFi is the poster child because of its high visibility. MetroFi is mentioned once in the release, while MetroFi's cities are called SkyPilot's "significant wins."
Joint Venture Silicon Valley has posted redacted versions of the WSV responses: Seven companies' initial responses were accepted for consideration for further winnowing by the JVSV group. The RFP covers 1,500 square miles and many municipalities. EarthLink's Don Berryman told me yesterday--as the company says in a letter declining to bid--that not enough homes are passed, there's too much "free" involved, and there's no streamlined process to deploy service in each participating city and town. I wrote several days ago about how I thought only one of the seven firms that did submit RFP responses could fund a network of this complexity and scale: the IBM, Cisco, Seakay, and Azulstar consortium.
I've looked through the RFP responses, and I stand by my original statement. That consortium is the only one with the apparent range of experience, current expertise, and easy access to funding that would allow this network to be built per the RFP (which might not be the ideal network, of course). MetroFi is a close second, but its lack of experience in working with multi-band public safety is a huge downside; public safety bands aren't mentioned in the RFP. They also would need a huge amount of capital to develop the plan, and the jury is out on ad-supported networks. While fee-based networks on this scale will also require deployment to figure out the market's value on them, it's easier to raise money when revenue comes from subscriber projections.
MetroFi would also have a difficulty in reselling its service. On page 35 of their RFP response, they note that their system "can support multiple service providers and content providers." But it's hard to imagine the same kind of ecosystem that EarthLink might generate for retail ISP brands in competition with their own when MetroFi is offering free service.
Community Wireless has an interesting proposal, but I believe it's too modest and limited to meet the spec. However, it's a realistic assessment of how to build a network of this scale without overspending.
Here's a quick summary of the RFP responses, most of which run into dozens of pages. (If you're looking to understand the current planning market and what equipment is in use, read these RFPs.)
Michael Oh is the force behind NewburyOpen.Net, one of the earliest and most influential commercially supported free hot zones: Oh started NewburyOpen.Net out of his computer consulting and sales firm, TechSuperPowers on Newbury Street in Boston. Experimenting with early AirPort gear, he was amazed by the distance signals would carry outdoors. His neighbors on the street became interested as he discussed the potential of sharing Internet access to their customers, and several businesses signed on.
The project turned into something bigger because of the prominence of that commercial street's access because Oh has evangelized the idea, writing detailed white papers that explain how to duplicate his efforts socially and commercially. SalemOpen.Net is a "franchise" of sorts in that Massachusetts city.
Oh's interest has broadened from Internet access into the power of the local network. In this podcast, we talk about the history and function of NewburyOpen.Net, and then move into how a fast local network can deliver content in a way that an Internet feed--with speed as a gating factor--can't match today. That disparity in local wireless network speed and Internet pipe will only continue to grow, too. [37 min., 18 MB, MP3]
Don Berryman heads EarthLink's municipal networks division: In this podcast, we speak about a huge array of topics centered around metro-scale deployments. How EarthLink, despite pricing and regulations, has a pile of existing broadband customers, and how they've been working on new broadband technologies for four years. Why Berryman says the numbers don't work for large-scale advertising-supported networks. Why early deployments have sparse coverage and have had to add nodes. How EarthLink's revenue model isn't dependent on Wi-Fi but also includes a substantial line item of revenue from their "capacity injection" layer, which can offer 1.5 Mbps to 3 Mbps via Motorola Canopy today on a point-to-multipoint link for business and government.
Other topics: Broadband over powerline (BPL) as a way to reach service to tall buildings (and EarthLink's investment in Current, a BPL tech/deployment firm); why EarthLink declined to bid on Wireless Silicon Valley; and why Berryman is so excited about the train-based Internet/operations network proposal for the Capitol Corridor rail line in California for which EarthLink is one of four firms building test networks. Whew. [42 min., 20 MB, MP3]
I'm still trying to sort this out, but it looks like Haier has signed on to using 802.11n in newer televisions with ultrawideband (UWB) nowhere in sight: In 2004 and then again mid-2005, putative UWB chipmaker Freescale trumpeted test with Chinese electronics and white wares' giant Haier. Last year, they were talking about how Haier would ship an LCD HDTV with UWB embedded, paired with a digital media server with the same wireless technology. These were "expected to be available" in Chinese retail markets in fourth quarter 2005 and in the US in 2006. I can't find any mention of this product having shipped, which isn't surprising given that Freescale partners have not to date shipped a single product bearing the company's UWB chips.
Meanwhile, Metalink announced today that Haier had demonstrated the use of the company's Draft 802.11n chipset in Haier HDTVs; they showed this at the SINOCES show this last week in Qingdao, China. Of course, this is an even more premature announcement than Freescale's last year, because this release doesn't mention any devices that would work with the TV sets, nor is 802.11n anywhere near done, and I doubt a consumer electronics maker would embed a draft standard. Update: Metalink said via email that Metalink and Haier announced in March of this year cooperation on media servers and similar consumer electronics products.
I'm waiting for word from Freescale as to the status of the Haier deal. UWB has seemed the ideal technology for consumer electronics where you want high speeds and very little configuration over short distances. Pair UWB with 802.11n for streaming content around a home as necessary, and you have a great distribution system. UWB and 802.11n have distinct purposes, so it's particularly interesting to see Haier flirting with both or changing dance partners.
Boston Globe article looks at Wi-Fi loiterers: A local angle to a story that I wrote a year ago April about a non-trend, but an interesting development, in cafes and other establishments that were restricting Wi-Fi use, sometimes turning it off for hours or on weekends. In Boston, there appears to be a lot more direct action than here in Seattle. We're passive aggressive; they may just be..aggressive.
An interview with Jim Baller, an advocate--literally--for municipal wireless: Baller is part of The Baller Herbst Law Group (Washington, D.C., and Minneapolis), and has worked for many years representing municipalities and public utilities in their efforts to offer services, particularly broadband. This involves negotiation, not necessarily litigation.
We talked today about a bit of history, starting with electrification 100 years ago, through the Pennsylvania and Philadelphia saga of 2004, finishing up with the Net Neutrality's current drafts of laws in Congress that include provisions to level the playing fields for municipal broadband. Baller notes that the versions before Congress don't tip the balance to towns and cities; rather, they require municipal entities to go through a reasonable process before building networks [36 min., 17 MB, MP3].
Welcome the latest addition to the Wi-Fi Networking News set of sites: You might notice a small change in the list of six sites at the top of this page--WNN Europe has been replaced with Public Safety (Wireless) News. The new site will be devoted to the rapidly emerging category of gear and networks that are used for first responders: fire, police, rescue, and specialized responders. Many municipal wireless networks are already in use that operate exclusively for public safety purposes; public access doesn't exist or is a distant second item. More networks will have the dual purpose, with public safety having priority.
WNN Europe is gone but not forgotten. The archives will remain active. In the time that site has been launched, we've posted under 200 items, and it's been increasingly clear in recent months that there are few Europe-only stories, but rather stories that span other categories.
Pittsburgh builds downtown Wi-Fi zone: The Pittsburgh Downtown Partnership will offer two hours of free access a day across a downtown service area. US Wireless Online will install the system--for which they will be paid $356,000 to put in place and run--and charge $7.99 a day to $119.99 a year for unlimited access. The service will reach into low-income neighborhoods. Cities in Pennsylvania can't install broadband without petitioning incumbent telecom firms, but this is an outsourced project and has little government involvement. (The Pittsburgh Post-Gazette adds the fact that US Wireless is reorganizing and considering a buyout offer.)
Uniontown adds free hotspot: A local firm, American Broadband, has put free service in Storey Square.
Bloomberg says yes, Microsoft no: Reports yesterday were that Microsoft was planning a Christmas release of an "iPod killer" to compete with Apple's dominant music-playing device, with Microsoft's offering Wi-Fi file transfer without a computer being involved. Microsoft said today that they didn't demo anything, but didn't say that nothing was demonstrated.
Meanwhile, you can read about the Microsoft Windows Media Player 11 Portable Media Player Device Media Player Media Player Media Player at Crazy Apple Rumors Site.
Minneapolis finally launches test networks: The finalists were announced last fall for building a Minneapolis network with fiber and Wi-Fi components, and US Internet and EarthLink will launch test wireless networks this Saturday. The city expects to announce a winning bidder Aug. 14, with a contract to be voted on Aug. 18 by the city council. That seems rather remarkable unless the contracts are already fully negotiated and ready to go on award of the bid. The city's deputy CIO says EarthLink discovered they needed 30 percent more Wi-Fi nodes than initially planned for the their one-square-mile test network.
New York starts to plan for a plan: On July 21, the city should receive proposals for conducting a broadband feasibility study, which would assess the city's existing facilities and whether a citywide network would be needed. Wired or wireless isn't clearly specified in this article.
The second acquisition in the last several months of a significant firm in the AAA (authentication, authorization, accounting) space: Meetinghouse and Funk were independent powerhouses, challenging Microsoft, Cisco, and others domination of a critical area of the networking space with their standalone servers and custom client packages. Both companies were active in the standards process, and I can't tell which was more important in EAP-TTLS, a secure 802.1X authentication method for wired and wireless network access that represents a strong alternative to the Microsoft and Cisco backed (and incompatible) flavors of PEAP. Both companies offered a radical expansion of client options to platforms that otherwise were unserved or had poor built-in supplicants for 802.1X.
Funk was bought by Juniper Networks in a deal closed last December; Cisco now buys Meetinghouse. This is a great deal for Cisco, which already makes a variety of RADIUS and AAA tools, because it lets them answer complaints among customers about choice and configuration by presenting a mature product line that will work perfectly fine with their existing client and server software.
The last remaining firm that I'm aware of at any reasonable scale is Open System Consultants down in Australia which make the Radiator Radius software, a well-liked package that requires some kind of brain expansion tool to configure. I tried, and got it working, but it's the kind of product that needs true dedication to unleash its seemingly bottomless potential, to judge by its manual.
Over at GigaOm, they're saying EarthLink will sell Wi-Fi-only VoIP phones in fourth quarter: For $10 to $25 per month, the voice plan will include calling and exclude data. They're testing phones right now for deployment, Katie Fehrenbacher writes, and expect the phones to work in transit--she reports their VP of VoIP says the testing shows current models work at up to 40 mph. Wi-Fi operators and aggregators have been settling on a much lower rate, such as the Boingo/Skype mash-up (still in beta after many moons and with hard-to-find, out-of-date copy) that offers unlimited service for $7.95 per month.
The biggest problem appears to be the cost of these phones--EarthLink will have to subsidize the phones for a while, and Fehrenbacher quotes VP Steve Howe saying, "this business will get really interesting when the phones get down to the $40 range." It's true. A new Linksys phone has all the right bells and whistles for using Wi-Fi on arbitrary networks, including a micro-browser for authentication and gateway page clicking, but it's nearly $400.
EarthLink caught me up on two issues, one foreground, one background: The recent announcement of the seven preliminary bidders for the Wireless Silicon Valley request for proposals (RFP) floated by the Joint Venture Silicon Valley public interest group included EarthLink's name, but not as a bidder. Rather, EarthLink declined to participate. In an article a few days ago, I speculated that perhaps it would require too much capital and offer too little return. Not quite so, said Jerry Grasso, EarthLink's spokesperson on municipal issues, in an interview this morning.
Rather, Grasso said, the Wireless S.V. bid would have required far too much free with far too few homes passed for service within the EarthLink financial model. "When you do the math, what the value wanted didn't fit with what EarthLink looks for from its business perspective," said Grasso. He said that reports (including a speculation on my part) that the company was overextended aren't true. The municipal division outsources the heavy lifting to partners, like Motorola's infrastructure building services division.
On a less public issue, EarthLink's Cole Reinwand, vice president of product strategy and marketing, answered a reader question--Will EarthLink offer wholesale access to other retail brands in their metro-scale deployments not just to the Wi-Fi network, but to the backbone WiMax-like point-to-multipoint business-grade wireless network? The answer is yes.
Reinwand said via email that they expect a large percentage of what they term "T1 Alternative" accounts to be sold on a wholesale basis to integrators who can handle the on-site installation of equipment, customer interaction, and sales and marketing to the smaller businesses that would benefit from this service.
Reinwand said EarthLink would handle sales directly to municipalities; in Philadelphia, several hundred leased wired lines will be replaced with EarthLink's backbone T1 Alternative at a much lower cost. EarthLink is using Motorola's somewhat ridiculously named MOTOwi4 systems for backhaul, which use Canopy (currently non-WiMax). Motorola might stick with the funny names for consumer brands.
Broadband wireless through point-to-multipoint has some very high advantages, exploited by TowerStream and other early movers, including fast install (sometimes within 1 to 2 days) and simple increases beyond T1 rates of about 1.5 Mbps with no additional equipment.
I have to eat my words of a few days ago, in part: The vendor who has the contract to unwire several New York City parks will finally get Central Park and others lit up. A few days ago, I wrote that I expected a July deadline to pass given the scope of what Wi-Fi Salon had yet to do and their apparent failure since Oct. 2004 to secure sponsors of the scale that would allow them to provide free service in the parks. They've signed up Nokia as the anchor, The New York Times reports, although it sounds like a near thing.
Here's why I say "In part." I always assumed that the Parks Department contract required service that would reach across the entirety of each park. This article says that 18 locations among 10 parks will be activated--eight in Central Park alone--which is a far cry from full coverage would which make mobile devices fully useful rather than just laptops.
The article points to Bryant Park as a success with 250 daily users during summer months.
Centrino: Cometa. Rosedale: Clearwire. And so on: May I be forgiven for invoking the late and largely unlamented Cometa as Intel Capital stuffs $600m (of a $900m total investment) into Craig McCaw's already bulging pockets? Absolutely, I hear you cry.
In July 2002, we first heard about "Project Rainbow," which was launched as Cometa that December; the company said they'd build 20,000 hotspots over two years, though the number was later revised down. The initiative was lauded as a combination of Intel, IBM, AT&T, and two venture capital firms. In reality, as I have tried to remind people ever since, it was Intel Capital (funding), IBM's service division (installation), and AT&T's broadband division (service). None of the companies had anything on the line when Cometa failed.
Intel hopped on Wi-Fi after the train had left the station. By the time they launched Centrino and spent tens of millions on a Centro-verified hotspot branding program, 802.11b was already clearly on the way out. Centrino systems wouldn't get 802.11g for another year, and Wi-Fi Protected Access (WPA) security lagged longer with upgrades from Intel that conflicted with manufacturers' Windows XP editions.
Centrino was a huge success from a branding perspective, and Centrino laptops had substantially better battery life while not offering great performance.
Cometa ultimately went belly up in May 2004 after securing just a few hundred hotspots, with a focus in Seattle, and losing the McDonald's bid to Wayport. The Barnes & Noble contract that they had signed just before shutting down was taken up by SBC--now called AT&T. Poetic justice of a sort.
The company suffered from setting unmeetable goals. Their founder, Larry Brilliant, now the head of Google.org, made statements that went far beyond what was achievable in a maturing market, and then left to focus on his hot-zone--as in infectious disease--specialty, leaving the company holding that that bag. Intel had also lost interest in Wi-Fi. It had become an also-ran to WiMax. Wi-Fi had no strategic advantage to the firm given that it was a de rigeur part of every modern laptop.
And that's why Clearwire, my dear friends, is no Cometa.
First, Intel has staked the future of its telecommunications effort on WiMax. They've pushed money into dozens of companies. They've headed standards bodies, built chips, and subsidized a lot of early network development. They've pissed off cellular operators by creating buzz around technology that clearly competes and complements cell networks, revealing some of the paucity of user-oriented thinking in the cell industry. Intel has burned bridges and sown salts in other fields. If they can't make WiMax bloom, heads will roll from the top down.
Second, Intel "owns" WiMax in a way that they didn't with Wi-Fi. Intel didn't make its own Wi-Fi chips and wasn't a leader in the market, which was already four years into development when Centrino shipped. While there are dozens of companies, including some enormous ones who compete with Intel on many fronts involved in with WiMax, Intel has consistently taken a position far ahead of the market and somewhat ahead of the industry. They're not leaving this standard to other people.
Third, Clearwire has a spectrum portfolio. No one can own Wi-Fi, but you can buy pieces of WiMax by having exclusive spectrum rights for regions of the country. Clearwire surely needs more licenses, and I'm sure has a strategy about that, but they can reach 90m people now, they said in their now-withdrawn IPO filing a few weeks ago.
Fourth, the technology isn't unproven or oversold by Intel or Clearwire. There is a lot of hype around WiMax, but I can't find statements in which Intel or Clearwire oversells how the technology will deliver. They're leaving that to other firms. If anything, the two companies are underselling WiMax before they can deliver the goods.
Fifth, Intel Capital has invested something like $1b in WiMax if you include this $600m in Clearwire investment. That's something like at least 50 times more than their Cometa investment.
Sixth, there's a clear market out there for fixed and mobile wireless broadband, which is why there are hundreds of wireless ISPs and billions spent on cell data. It's also why the metro-scale wireless market is so hot, and why Wi-Fi--despite it being not the ideal technology for the purpose--will be blanketing hundreds of cities by the end of 2007.
Cometa had nothing unique over other hotspot operators except a staff split among at least three different headquarters that was much too large and a lack of signed contracts by the time they shut down. Hotspots were on the way to becoming a standard commodity that could be built and resold by a host of firms even as Cometa launched.
Clearwire has a particular set of advantages. They have Intel's attention. They have a chunk of spectrum licenses. They have a technology that works. And they have an audience that wants truly large-area mobile service from anyone but the incumbents--even if that's just for competitive sake to drive prices down.
Clearwire will execute, there's no doubt, and the question will be whether they can acquire enough customers at the right price to turn a profit. Cometa once talked about creating hotspots to such a large extent that they would cover every area in which a business traveler might find themselves. Clearwire has the potential to meet that target and avoid Cometa's fate.
Craig McCaw's broadband wireless firm Clearwire raises $600m from Intel Capital, $300m from others: The latest revolutionary wireless firm founded by McCaw aims to deploy mobile broadband wireless worldwide using mobile WiMax (part of 802.16e-2005). Part of the money comes from Motorola purchasing Clearwire's NextNet equipment subsidiary, which has been manufacturing and prototyping gear for Clearwire's network, starting with customer premises equipment (CPE), or the fixed receivers plugged in at homes.
Clearwire owns the second-largest portfolio of spectrum in the desirable 2.5 GHz band in the U.S.; Sprint Nextel is the biggest holder. This is a great band into which to deploy mobile WiMax because of the geographic coverage--Clearwire says that they can reach 90m residents with current licenses--and the channelization, which is wide enough to allow sufficient bandwidth for real mobile applications, including video. (While BellSouth owns a chunk in 2.5 GHz, their biggest holdings are in 2.3 GHz. They are already looking at equipment that would offer WiMax or WiMax-like services in both bands. This spectrum is part of AT&T-formerly-SBC's desire to purchase BellSouth, which would also give AT&T 100-percent ownership of Cingular, and allow more combined offerings there across DSL, cell data/3G, and WiMax.)
Intel has had a chicken-and-egg problem with its backing of WiMax, particularly the mobile and portable/nomadic form, in that they need networks to drive interest in the chips they plan to include in their laptop reference designs. By investing this heavily in Clearwire, they've basically guaranteed that a network will be built. This also seeds more interest in competing networks, and puts the cellular operators on notice that Intel is not their partner, if they ever harbored such a suspicion. In fact, Clearwire could offer competitive voice services over their network using handsets with mobile WiMax built in.
Intel is slated to ship Rosedale 2 chips by the end of the year, according to Light Reading, which will offer both older fixed (802.16-2004) and newer fixed/portable/mobile (802.16-2005) support. They'll also make Ofer-R available for Wi-Fi and WiMax support in portable and handheld devices.
Way back at the Centrino introduction, Intel told me that future Centrino wireless chipsets would incorporate Wi-Fi and cellular data standards. That never happened. Instead, Intel discovered the wonders of a newly competitive marketplace that they thought could evolve worldwide in which they could have a stake and a say in its operation and standards development. Intel has been a big force in WiMax from many angles, this being just the latest.
New York Newsday makes the Parks Department's ultimatum to its contractor sound like a firm plan: In May, the city that never sleeps (or unplugs) told its winning bidder for the rights to install Wi-Fi in several major parks that they had to get Central Park unwired by July or lose their contract. They also said that the fees required for the rights would be reduced or removed from future contracts, as that was impeding these rollouts.
Newsday's coverage has no new information, but takes it as a fait accompli that the park will have coverage. They have no quotes from nor mention any attempt to contact the contractor, WiFi Salon, which has had these contracts since Oct. 2004, and has always said they'd use sponsorship to offer free access.
Dana Spiegel of community wireless group nycwireless points out that the article reveals the Bryant Park service cost $18,000 to setup and $1,400 per month to run.
Unless I misunderstand what's happening there, I expect we will see a different headline in August than, "Central Park Wi-Fi up and running." If anyone knows differently, please post in the comments and explain how you know.
Airport and hospitality operator ICOA loses its chief executive of 18 months: Rick Schiffmann along with chief financial officer Stephen Cummings had their resignations accepted by the board June 26 according to an SEC filing I was alerted to. Part of the filings explain that Schiffmann and Cummings have several months of unpaid salary. ICOA agrees as part of the separation to pay back wages and other fees, reprice options for Cummings, and agree to strict terms under which default could be declared. (Update: On July 6, 2006, ICOA put out a press release in which the company states it requested the resignations. The SEC filings state simply that resignations were accepted.)
As I noted in a story in April after reading the pink-sheet traded firm's 10-KSB filing, the company stated they need to raise at least $2.2m in financing to continue its expansion. In their list of provisos about the company's finances--under the heading "Going Concern"--the firm said, "The ability of ICOA to continue as a going concern
is dependent on ICOA's ability to raise additional debt or capital...We remain confident of our strategic direction and recognize that the investment in growing our business will take some time to generate sufficient cash-flow to support all of our operations and opportunities. Strengthening the Company's cash position remains a top priority for 2006."
This filing, as far as I can tell, mentions nothing about deferred salary.
Schiffmann was a key figure at SoftNet, where Ken Brilliant first tried his hand at Wi-Fi via Aerzone, and then founded his own venture, Airport Network Solutions. He then moved to Cometa, Brilliant's second Wi-Fi venture, for just a few months (Dec. 2002 to April 2003), according to his company bio. (Brilliant is now pursuing more noble ventures at Google.org.)
A little weekend news about a very large project: The Joint Venture Silicon Valley business group paired with a joint government coalition to issue an RFP a few months ago to unwire the entire valley. I wrote about the scope of this project in January, and the RFP was issued in April. The RFP specified a lot of purposes, from independent low-power sensors to public access to seamless high-speed transportation services. No particular technology was demanded. The number of municipalities involved is now 40, up from an initial, including many of the cities in Silicon Valley.
The Wireless Silicon Valley proposal couldn't possibly cost less than $50m to build, and I would imagine it could exceed $100m depending on how the RFP was answered. That's a lot of capital to come up with, and most RFPs require the financing be secured in some fashion before the final contracts are signed. It's certainly possible to get funding based on being a finalist in a deal this size, of course, but that's dependent on capital providers believing in the business model.
I'm afraid that I will put noses out of joint by stating that I believe only one of the firms or groups that responded to the RFP has the resources to fulfill it at this writing: the consortium of Azulstar, Cisco, IBM, and Seakay. This group also put a bid in for San Francisco, and it combines a lot of technical and operational expertise with a public-service focus that was strong on privacy. Cisco and IBM can fund this out of pocket change and write off a losing proposition as good PR and experience.
Let me walk through why the others are non-starters for me. I can find no information about Blue Horizon Group or "Community Wireless" (a group's name or just a label?). Unless these two firms are well funded and have been running extremely quietly, they don't have practical experience in this sphere. Fire2Wire/Ubiquity Broadband Communities appears to be a well-run Web hosting firm that's breaking into a new arena. They may be great, but I don't see practical experience on their Web site, either. Next WLAN Corp. makes equipment, and lists no practical network-building experience on their Web site. They may do private network building as well.
VeriLan at one point was a small wireless ISP in Portland. They split their business into an events-oriented one (VeriLan Event Services) and broadband wireless (VeriLan, Inc.). Confusing, the former is verilan.com and the latter verilan.net. They haven't built a metro-scale network to my knowledge--they don't mention even operating service in Portland on their Web site any more--although they bid on Portland. They used Vivato equipment at one point in their history, then upgraded to newer metro-scale gear.
MetroFi would seem to be a good candidate for this proposal at first glance, but I can't see how they could pull it off. They were a small firm until recently, and must be in the middle of raising at least $30m in private capital to build the city contracts they've won in recent weeks. They have a pile of bids and contracts out, too, that might require $10m to $20m more. Without having executed any of these larger deals, I'm not sure how forward-thinking the venture or capital markets are on a project of a scale that has never been done that must be built at the vendor's own expense, even though there's a lot of money for actual services that cities will shift their way. But you have to build the network first.
MetroFi also has no ostensible experience beyond primary residential service through Wi-Fi using a single vendor's gear--SkyPilot. The Wireless Silicon Valley proposal almost certainly requires a mix of different kinds of wireless data standards, and SkyPilot only encompasses some of this potential, while MetroFi hasn't built these other sorts of networks. Call them a dark horse based on these factors, and most likely one of two or three finalists.
The press release specifically notes that EarthLink declined to participate in writing. They're not the be-all and end-all of this industry, but it shows that either they didn't want to go to capital markets to get the money or felt they couldn't produce a return based on the RFP. They may simply have been overwhelmed, too, with the scope of what they have to do this year.
I was hoping to see some other consortiums, including non-profits combining with for-profits, and I thought some of the larger telcos might have been involved, even to the extent that, say, Verizon or Cingular would have pitched rolling out EVDO Rev. A or newer HSDPA on a faster timetable in the area coupled with fiber-linked Wi-Fi in some areas.
It's possible the Azulstar/Cisco/IBM/SeaKay group will simply be judged to have the necessary experience and funding to make this happen and get anointed. The Wireless Silicon Valley press release says that this list of seven will be winnowed down to a short list, and then the best proposal or proposals, they note, would be announced in September, and then, I would guess, trial networks.