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.
The head of the non-profit that manages Philadelphia's relationship with EarthLink says the goal is empowering the Internet-less: Greg Goldman writes in an op-ed piece in the Philadelphia Inquirer that metro-scale Wi-Fi networks are "s about connecting families that have been cut off from technological opportunity, and creating exciting new opportunities for everyone." I'm not sure what the translates into in practical terms: In what ways are their lives better? It's a short op-ed, so he's been required to be brief.
He writes, "We recognize that pressure to restructure municipal wi-fi business models may be mounting, but we urge that the long-term potential of these initiatives not be sacrificed to short-term objectives." Sure. And somebody needs to foot the bill. Since networks now won't be built by using a money cannon to shoot dollar bills at utility poles, cities may need to step up to the plate if there's a translatable benefit to society.
Would it be cheaper to solicit alternative telcos to install DSLAMs (DSL aggregators) in less-served central phone offices, providing those telcos with guaranteed rates of return for investing in areas with traditionally low levels of subscriptions? I don't know, and I imagine that the point isn't the medium. In this case, Wi-Fi is less encumbered (despite present difficulties), and that should have led to faster rollouts than equivalent wireline means. It didn't, so maybe some rethink is in order.
Despite years of bad regulation and court rulings that gutted most competitive DSL providers, alternative providers do exist and they can install their own gear in central offices. It's just bloody hard for them to make money. So the dollars involved in subsidizing real wired broadband is worth thinking about now.
Also, there's the Meraki model. What if neighborhoods bought bulk bandwidth in the form of a few expensive broadband lines (assuming they're expensive--they may not be) and shared it throughout the neighborhood? Wouldn't that be cheaper, more immediate, and more effective? That's what's being discovered so far.
And a nit to pick with Mr. Goldman, something I addressed with Wireless Philadelphia via an email exchange about this topic. "Subscriptions to the network have more than tripled in the last quarter," he writes. Tripled from what? EarthLink is currently dedicated to denigrating their muni networks, and hasn't released usage numbers in a while. A little birdie told me that subscribe numbers in Philadelphia in covered areas are quite good. Goldman confirms that--but won't mention the numbers, apparently. This may be due to agreements with EarthLink over business metrics.
Still--tripled from what?
I am remiss in posting a link to Google's peek under the kimono for usage of its free Mountain View network: The network that Google was able to build, a little late, but generally to good reviews, sees a lot of usage, the company's "alternative access" team posted to the Google blog last week. (Yes, I'm late to the game on this one.)
They have over 400 routers covering 12 square miles, which conforms to the general expectation these days of needing about 35 to 40 routers per square mile. They see 15,000 unique users each month, although we don't know with their methodology whether that's uniquely created accounts, since you can create disposable accounts on their system. They transfer 300 GB of data each day to 100 different Wi-Fi devices. Given the different number of adapters, that's the iPhone, 95 laptop cards, and four other devices (I'm joking).
Also interesting is that 95 percent of their mesh nodes see some use each day, showing that it's not just concentrated usage. I'd love to see a histogram of usage, showing the amount of usage against the percentage of routers.
Will Google build more Wi-Fi networks? No. But I can see them putting money in different forms into new models of city-wide Wi-Fi that are already underway.
AT&T hasn't gotten the first phase of St. Louis, Mo.'s network running due to pole problems: Light poles on which the telecom giant planned to place Wi-Fi nodes are largely bank switched, which means that lights are turned on at night, and there's no circuit at the light to control illumination separately from other devices. Light poles in some cities are wired to allow the light to be separately controlled from other devices, or integrate switching at the pole. They've spent months on the problem, with no solution. The project might be scuttled.
A bit more reporting from Houston, more negative than yesterday's: Sounds like there is more skepticism yesterday than the day before about whether EarthLink will build that city's network (including a comment from yours truly). The city could use the $5m payment from EarthLink for missing its first milestone--money due in a week or two--to build hotzones.
I had a dream last night that I was on EarthLink's board: I went to a distant city where we had outsourced the construction of a city-wide network to a local group. When I arrived, I found the network performed extremely poorly; really, hardly at all. I went to confront management, and discovered that the company was actually a group of American-flavored religious extremists. They held me captive for two days until I escaped--rather easily, they weren't very clever--with my dream-only dog to return to EarthLink HQ. News accounts of my escape hadn't mentioned my name or position. I sat down with the board and said, "I was just a hostage for two days." And then we proceeded to talk about the network's future.
My article on municipal Wi-Fi appears in the next issue of the Economist, online now: Those who read this site regularly know my take on the matter, which is that overpromising of indoor Wi-Fi coverage by some firms and cities (not most, thankfully); an expectation of much higher subscriber numbers; underbuilding of the network in early phases based on overoptimistic coverage estimates; an rush to the bottom to provide no-cost networks to city with revenue chasing the subscribers who never arrived; and a lack of plans by most cities about what, in particular, to do with the networks that were being built has led to the current implosion.
While some folks in the industry have tried to persuade me that public safety money won't drive the future of metro networks, I find that as I talk to cities and counties--often dozens a month through casual email or more in-depth conversations--that unwiring public safety is a high priority. The 4.9 GHz band is really taking off, and the availability of equipment that's starting to become commodity priced for 4.9 GHz receivers and adapters makes it a good move for cities and counties from typically proprietary gear running on narrow slivers of bandwidth. (That's the whole Sprint Nextel spectrum reallocation nonsense still in play.) Not all communications are suited for 4.9 GHz, but a lot of critical ones work quite well, including video and high-bandwidth transfers from officers and workers in the field.
Combine public safety money with municipal needs as they better articulate them, and you have a business plan in which a service provider's first and foremost set of numbers deals with providing a fixed set of services at a given rate of profit. Adding service to the public for mobile and, as 802.11n starts to penetrate, some indoor access becomes an incremental revenue source above the baseline that the provider establishes to build and operate the networks.
As Chuck Haas of MetroFi point out to me, if a city wants to have a 4.9 GHz public safety network, there can be only one client and one line of revenue: the city. Thus, 4.9 GHz networks must be fully funded with the profit motive in mind. That provides the base on which 2.4 GHz networks could be built.
EarthLink wouldn't comment to the AP, but Gavin Newsom's office says deal is dead: The Associated Press is reporting that 19 months after they were awarded the bid, EarthLink is out. I called it several weeks ago: the new model made it clear that there's no way for San Francisco's desire and EarthLink's approach to meld now. The network was estimated at $14m to $17m to build, and Google would have been an anchor tenant, paying EarthLink to provide a free, lower-speed (300 Kbps) service.
This isn't the beginning or even middle of the end. This is the end of a failed approach in which risks, dollars, and expectations weren't in line between municipalities and service providers. This was the last axe to fall on major deployments. In six months, we'll start hearing the good news about how realigned goals and more reasonable network plans produce results.
Hardly the best way to kick off expense cutting: A day after EarthLink announced layoffs of nearly half its staff, including the head of municipal networks, the company has agreed to pay Houston $5m because of delays in starting its network construction. The city announced this payment this morning, and EarthLink now has nine months to begin construction. That $5m would be twice the city's $2.5m anchor tenancy commitment. The city may also consider other offers now, too, without any obligation to return the money.
I've been saying for weeks that I thought it unlikely EarthLink would build out Houston in the end. Paying a $5m penalty and walking away is cheaper than building and operating a $50m network.
Springfield, Ill., has plug pulled by AT&T: AT&T said that it wouldn't deploy free service through the capital of the state. The state might still provide money for a free downtown and statehouse hotzone. The city has already met with a new wireless entrant Xanadoo, which is offering Navini's flavor of WiMax technology. Xanadoo has 2.5 GHz licenses covering 8m people in several states, including Illinois, and 700 MHz spectrum covering 170m people in major markets, according to a June 2007 press release.
Houston up? Houston down? The Houston Chronicle reports that EarthLink wouldn't comment on the future of the Houston network after the companies layoffs, where Houston is an anchor tenant, but is paying just $2.5m over five years for a network estimated to cost $50m to build. The city said they'd hold EarthLink to penalties if the contract weren't fulfilled. A utility pole agreement still needs to be signed and is overdue. The city told the Houston Business Journal, however, that EarthLink has committed to the mayor to build the network, but needed some time to get the company realigned before moving forward. Update: EarthLink pays $5m penalty; Houston can shop around.
Colorado cities agree on provider: 10 cities around Denver agreed to a common set of goals for a wireless network for economic development, emergency response, education, public safety, and visitors. There are 620,000 residents across 137 square miles. C-Com was selected to proceed into negotiations; they'll use BelAir gear. This network will allow public safety and municipal workers across the 10 centers to have access to a seamless network, too, which should aid in efficiency for shared services and tracking down criminals. The cities expect to pay for services, and estimated that they spend $4m collectively on voice and data; they would like some measure of free service on some basis to residents. Interestingly, C-Com wasn't on the list of respondees as of the close of the RFP process (EarthLink, Kite, and MetroFi were). Their site wasn't updated at this writing to include the release and information sent to reporters.
EarthLink CEO elaborates on layoffs, plans in call this morning: In a media/analyst conference call this morning, EarthLink's CEO Rolla Huff said that the municipal Wi-Fi business isn't going way. He said the company had scaled staff and resources ahead of potential, and for a few lines of business "we are scaling back their cost structures to make them the startups that they in fact are." He also said in regards to networks like Philadelphia, they will scale those networks "if we believe we can get to scale that covers our operating costs." Which implies that Philadelphia could still not be fully built out.
He also said intriguingly that because there are several constituencies which have a "vested interest in seeing municipal Wi-Fi networks exist"--which he listed as "municipalities, chipset makers, equipment manufacturers, or, yeah, even WiMax providers"--that "there could be an interest in a broader sharing of the costs." Does this mean that Huff has been on the phone with Sprint Nextel, Clearwire, Cisco's Linksys division, Intel, Apple, and others? Almost certainly Sprint Nextel and Clearwire; the other firms, less clear. It could make a lot of sense for WiMax providers to have a stake in Wi-Fi networks that aren't competitive but complementary.
Huff also reiterated the the model in which EarthLink fronts all the costs is "simply unworkable." Which we'd heard before, but this was a bit stronger than last time around.
The ferries are unwired, just not all of them: The Seattle Post-Intelligencer reports that the usual suspects of location, location, location are bedeviling the addition of Wi-Fi-based Internet access to several ferry runs in Washington State. The ferry system in my home state carries 50 percent of the ferry passenger trips in the U.S. A good half of its routes by traveler numbers have had Wi-Fi from some time, supplied by Parsons.
But it's the last part that's problematic. Siting issues have held up placement of antennas, which in turn led to trickier engineering tasks. Parsons said it was optimistic in its deadlines for adding Wi-Fi, thinking they were ahead of the game; the executive in charge of the project says ruefully they should have stuck with contracted target dates publicly and then been happy if they'd beaten the marks.
The reporter rode some ferries to talk to passengers to see if they were missing the Wi-Fi access. None were. The article concludes with a bit of poetry, following one passenger's lament about the cost of the service, which was too high for his casual uses of email and YouTube: "And there was nothing left to watch on the way home but the shimmering blue water and the seagulls flying overhead."
It's a big day in the future of metro-scale networking: It's not a crash, but it's a big readjustment.
Chicago cancels Wi-Fi plans: After months of negotiation with bidders AT&T and EarthLink--those are separate, not joint bidders--Chicago decided to cancel its Wi-Fi plans, the Chicago Tribune reports. The city and firms couldn't come to terms. The Tribune says the city wouldn't offer any anchor tenancy commitment; AT&T and EarthLink required that. I can see the meetings. "Free." "No." "Unencumbered?" "No." "No cost?" "No." And so on. Chicago will re-evaluate how it goes forward. AT&T meanwhile now charges $20 per month for 1.5 Mbps ADSL, as well as their merger-required $10 per month/768 Kbps offering. (Ten percent of Chicago residents can't get DSL, but the cable penetration for that 10 percent isn't noted.)
EarthLink lays off 900, municipal division head out: Donald Berryman, head of the company's Municipal Networks division is out--his "employment with the Company is terminating," reads the SEC filing (PDF)--and the position he occupied as an executive VP and president of the division is also gone. 899 of his colleagues at the company are also being laid off. It's unclear how many others, if any, come from the municipal division, but I can't imagine they'll shed the head and that many other employees without some attrition there. The municipal group outsources construction; I don't know it's current size.
EarthLink expects to save $25m to $35m through 2007, although they'll have one-time charges of $60m to $70m. Savings should exceeds those costs within six months. One city in which they will "substantially reduce [their] presence" is Atlanta, where they're headquartered. Companies that change CEOs sometimes change their HQ cities, too.
Useful to note that the new CEO gets a variety of stock incentives, including 700,000 options that vest Sept. 30 of this year, and 800,000 vesting starting Dec. 31, 2008. He was also awarded restricted stock that doesn't vest until starting June 25, 2009. Now, I thought giving executive the desire to make short-term gains was something of the past? Huff could pocket $7m for a $10 stock gain from options over a few months based on strong moves that may have poor long-term revenue consequences (or good ones). The other incentives are longer-term, but he doesn't have to stick around for those. The down side for him is that he did purchase 100,000 shares with his own pennies and dollars (about $750K), so should his efforts fail, he's underwater on EarthLink options and stock as well as out of pocket. Huff just hired a crony, too; the new COO, Joseph Wetzel, was the president of operations at Huff's previous employer.
Speaking of Atlanta, what's the status of its Earthlink-winning Wi-Fi network?
Kite Networks deal goes south: In July, MobilePro sold its Kite Networks division, that runs Wi-Fi networks in the Southwest, and other forms of broadband wireless networks elsewhere--including some under the Sprint name--to Gobility. The deal was for future stock (convertible debentures), and required Gobility to raise $3m in financing by mid-August. An SEC filing from MobilePro states that Gobility failed to raise the funds and let some vendor payments for Kite Networks go slack. MobilePro says it may sue based on alleged misrepresentations from Gobility. Given how little money was involved in the Kite deal, it's unclear how much more MobilePro is willing to invest in its ventures. They provide Wi-Fi service in Tempe and part of Chandler, Ariz. Update on Wednesday: Kite sent out a press release (not available online) in which Gobility's head says that they're still operating the networks and working on funding.
Unstrung reports based on one source that SkyPilot has allegedly slashed half its staff: The company wouldn't respond to a query from Unstrung about what they told that outlet were "rumors." Unstrung says their source said half the company's 80 employees were let go, mostly in sales and marketing, and the company is looking for a buyer. SkyPilot provides equipment for MetroFi's metro-scale installations. I have been wondering with recent reversals in metro-scale networking fortunes what happens to the equipment makers that rely on large-scale installations such as SkyPilot, Tropos, BelAir, and Strix? The market is still growing despite marquee dropouts, and enterprise sales are typically unreported in this context.
Google's home town network sees 15,000 monthly users: Almost every node in the network has a user every day.
Geneva adds Meshlinx service: Geneva, Ill., joins St. Charles in choosing Meshlinx for service. Batavia, the third city in the triad may also join in. The three cities were negotiating with MetroFi months ago, but opted out in April after the firm pressed its need for an anchor tenant service contract. MetroFi is unwiring Aurora, Ill., a pioneer in electrification in the U.S., but utility pole issues have poetically delayed the rollout. 512 Kbps service will be free; faster service at a fee. However, the company is stating they'll put out about 25 nodes per square mile, so we all know what that means (see an infinite number of preceding metro-scale Wi-Fi articles on this site).
Ultrawideband (UWB) chip shortage delays Lenovo laptop rollout: Or not. Lenovo says it doesn't have enough UWB chips to ship its laptops on time; WiQuest says, tuh-uh, they're supplying plenty. Strange communications problem.
Two more cities officially on hold: The two Virginian cities of Alexandria and Arlington were planning on being built out by EarthLink. Now--less likely. The contracts were in Dec. 2006 and April 2007, respectively. [link via MuniWireless]
The clear-the-inventory site Woot offers 30 GB Zune for $150: They might be out by now...but maybe not. There are three reasons for Microsoft to clear the shelves. First, they have a new model coming and want to empty the inventory. Second, they want more devices in more hands to help create mass. Third, they're killing the product. Since the first release of Zune absolutely suck suck sucked, let me just be frank, there's little reason to buy it at any price. Zune created such a vaccum around itself because it had among the best featureset but the worst enabling of such featureset. The iPhone is similarly limited in its use of Wi-Fi: no song buying, no computer syncing. At least the Zune let you "share" songs with friends, even though the problems with that were rife. The iPhone, however, is beautiful, useful, has a redundant network (EDGE + Wi-Fi), and is part of people's existing software expectations: it's like an iPod, so I sync it in a dock. Anyway, if price were the only reason to not buy a Zune, you don't have that excuse any more.
Latest sorry bastard arrested for kiping Wi-Fi: A fellow in Chiswick, West London (that makes me feel all Jane Austen-y to type it) was arrested when spotted using a laptop outside a house. The policy were called, he admitted to using unsecured Wi-Fi, and he's released while the computer crimes division investigates. The use of "dishonestly" obtained Internet access is a crime in the UK.
To understand what's wreaking havoc on your wireless networks, a spectrum analyzer is key: Sure, you can make guesses. Talk to neighbors. Use Wi-Fi and Bluetooth scanners. But those approaches have limits. What you really need is a spectrum analyzer that can scan the surrounding area across a frequency range and show numerically and graphically what's in the air around you. By moving around with a mobile analyzer, you can pinpoint actual problems, and see displayed over time a moving target as to what's destroying your network's utility.
MetaGeek has two affordable options for desktop Mac OS X, Windows, and Linux analysis, both of which come in the form of a USB stick. When I say affordable, I mean that their units retail for $199 and $399 versus $2,000 and up for full-featured IT management packages and hardware or standalone devices. The two options scan the 2.4 GHz band used for 802.11b/g and one of the bands 802.11n supports. Bluetooth, ZigBee, and cordless phones--among many other devices--also use the band.
The original Wi-Spy debuted over a year ago; the Wi-Spy 2.4x is a more recent entry from June of this year. The differences between the two units were originally software support, with Mac OS X software available just for the original WiSpy at the launch of the newer device. The Wi-Spy 2.4x now has two Mac OS X supporting applications, too, due in part to the programming interface provided by MetaGeek.
MetaGeek has a nice comparison table that explains the difference between the original and 2.4x versions of their product. It boils down to the 2.4x offering greater resolution--it can capture signal strength on a smaller set of frequencies at once--along with higher sensitivity, and an antenna. The antenna can be removed and replaced with others using the same jack style (RP-SMA).
I tested both units in my office under Windows Vista and Mac OS X. My office in Seattle's Fremont neighborhood, a mixed retail/office/residential area, has a plethora of Wi-Fi networks. I can see from eight to 14 networks in my office, which is nestled inside a building. I tried turning on a microwave oven adjacent to my office to see what effects that created, too.
Using MetaGeek's own Chanalyzer software, I could watch Bluetooth dancing all over the 2.4 GHz band. Chanalyzer lets you set mark points to track in planar view, in which activity is marked for average and maximum uses across the band, while a live line shows current signal strength. The topgraphic view shows a concentration of activity over time. The spectral view is a moving track that you can replay and change the time dimension on to expand or contract a moment you're viewing. The program lets you "record" a scan, which you can replay in the software. MetaGeek also created a community for sharing scans among those interested in that sort of thing, and has samples you can download and replay in Chanalyzer to show various tests they've performed. (Chanalyzer 2.1 works under Windows. WiSPY-Tools is an independent Linux, BSD, and Mac OS X package with similar features. MetaGeek has a beta of Chanalyzer for Mac that's not currently linked on their Web site.)
The scan above shows the normal environment--fairly congested, and you can see the Bluetooth spikes (by clicking to view the full-size image) as lines that leap up in yellow in the planar view. When I turned on the microwave oven, there was some impact (see at left). In the spectral view, you can see the diagonal green lines between 1 and 2 minutes on the time scale when the microwave was first active; I turned it on again briefly and you can see another set of angry lines. In the topographic view, notice the haze of interference rising above the deeper, thicker red band.
I also experimented with EaKiu, a free and independent Mac OS X software package that supports both the original Wi-Spy, and was updated recently to version 4.0 with Wi-Spy 2.4x support. The package offers some interesting 3D options for visualization, in which you can rotate a continually updating series of receding planes.
I honestly found it tricky to figure out how to test the Wi-Spy models since I had no particular issues facing my rather ugly RF environment at the office, but one fortunately dropped into my lap. Apple recently released its revised model of AirPort Extreme Base Station with Draft N by adding gigabit Ethernet. They also tuned some internal firmware issues to improve speed when network address translation (NAT) was in use. For a review in Macworld magazine, I re-ran the same performance tests that I put the Extreme through in February. In the 5 GHz band, everything was copacetic: terrifically improved speeds due to the gigabit Ethernet removing internal limitations that restricted performance.
But in 2.4 GHz, I was stymied. I could hardly get my N adapters (Intel and Apple) to connect to the base station. With Apple's advice, I turned off Automatic channel selection and chose channel 1. I was seeing kilobits per second throughput when I could connect where I should have seen 30 to 70 Mbps. I fired up the Wi-Spy 2.4x to see what was the matter.
It seemed like I was seeing a lot of energy down at the lower end of the band even when the Extreme wasn't transmitting. I switched to channel 11, and, magically, performance was restored. Now, according to iStumbler, no one is transmitting in channel 1, so there's other ugliness involved. If you look at the Eakiu screen capture (at right), you can see the time sequence moving from newest to oldest on the Z axis (roughly front to back). I switched from channel 1 to 11 right where you see energy levels go down and red (higher dBm signals) disappear on the left, and start to fire up into red on the right. Clearly, something's using that part of the 2.4 GHz band (amateur radio? electronic newsgathering? another licensed purpose? an ugly transmitter?). But without the Wi-Spy, I would have been slightly flummoxed.
The Wi-Spy isn't for every network manager or hobbyist, but it's going to help IT professionals and Wi-Fi busybodies like myself answer a lot of questions that are otherwise lost in the ether.
VIA Rail Canada has put Wi-Fi into Vancouver, Jasper, Edmonton, and Winnipeg: The rail line travels from Vancouver to Toronto over three days; Wi-Fi will allow some interim contact, apparently. The service is for-fee for regular passengers; free for first class. Rates are C$3.99 for 15 minutes (plus C$0.30 thereafter), C$8.95 for 24 hours, and C$46 per month (special rate of C$29.95 until Aug. 31). VIA Rail was among the first firms to add Internet access to trains on a production process. It's available on their Québec City-Windsor run.
Lexmark pushes out a passel of wireless printers: The models run from $130 to $200. The printers all handle two-sided printing, which, in Lexmark's words "saves paper and the environment." Whew, I was worried about the environment, and printing on both sides of a sheet of paper apparently saves it. Three of the models are multi-function printers (MFP in industry parlance), which print, copy, scan, and fax. Lexmark is calling them all-in-one (AIO), just to be different.
The X6570 ($150) has 28 ppm (page per minute) black and 24 ppm color printing (draft mode print numbers, naturally, a footnote explains). You can remotely control the printer to scan, fax, and read memory cards inserted into a media slot. There's a 25-page document feeder, too. The X7550 ($200) ups the rate to 30 ppm black and 27 ppm color while including an LCD display for photo preview via the media slot. The X4850 ($150) appears to omit the fax feature, although Lexmark is unclear about that.
Lexmark also added the Z1520, which is just a wireless color printer (30 ppm black, 27 ppm color) for $130. You can also get wireless-omitting versions of the AIO printers, too, that have various subtle differences to boot (the X5070, $90; and X5495, $100).
Interestingly, Lexmark only uses the terms 802.11b/g/n, and then only in a footnote. That makes me suspect that the printers haven't achieved Wi-Fi certification yet. They're not in the Wi-Fi Alliance's list of certified printers, although the previous X4500 series is listed. It's an odd thing today to see "Wi-Fi" products without the Wi-Fi brand.
Hey, Blomquist! 2005 called and it wants its essay back: The American.com, apparently a red-blooded site about and for 'mericans (read: a specific subset of big-business-oriented conservatives) suggests that asking cities to pay for services in a kind of [air quotes] public-private partnership [air quotes] is tantamount to socialism.
The essay writer, however, while au courant on some details surrounding EarthLink, appears to confuse the idea of a city paying a private corporation for services with a city building its own network in order to compete with wireline.
The public-private partnership notion exploded some time ago. Municipal networks are typically--not exclusively--a business relationship of the same time that cities pursue with cable and telephone companies. EarthLink and MetroFi's desire for city dollars isn't to be paid to build the network, which is what Blomquist implies even though he states the opposite. (Rather clever rhetorical device, actually.)
He states that EarthLink wants a service commitment to make "at least a minimum return," but then notes that service providers will use this as the thin edge of the wedge. "As municipal Wi-Fi companies become entrenched, we can expect them to not only demand that cities be anchor tenants, but that they help make up for the shortfalls that will inevitably result when new technologies supplant Wi-Fi."
He also casts the Wi-Fi providers in the form of a monopoly because of this logic:
"Companies that enter into public-private partnerships enjoy favorable treatment. Special access to city or county rights of way, shelter from liability, and the backing of a public partner with the power to tax makes Wi-Fi firms that enter into municipal deals formidable foes. Potential competitors may find these advantages too much to overcome, thus driving them out of the marketplace. With potential competitive rivals driven out, we are soon left with an ossifying public utility disguised as a private enterprise."
I don't know of any contract that shelters the provider from liability, nor any city of scale (above small towns) that has found the possibility of using taxes to pay for the network palatable. (St. Cloud, Flor., is the exception, and a change in Florida property tax law may doom that network.)
Blomquist is more accurately describing cable and telephone providers, who receive rights of way in long-term contracts, and have enshrined rights in the mere fact of having installed copper and fiber in the ground that insulate them (hey, good metaphor) from any form of competition. They own the wire and glass, and there's no necessity for them to resell it at any reasonable price or, in many cases, at any price whatsoever.
So I don't see how AT&T, for instance, has any "competitive rivals" except for equally monopolistic cable firms. Telecom and cable companies fight it out for market share, but they barely differentiate their services except that cable can currently offer more speed than DSL with less effort and cost. (That's why telecoms are pushing fiber to the node, curb, and home.)
What Blomquist really misses is that Wi-Fi has been shown to be a poor competitor against wireline services. Wi-Fi might be a dial-up replacement, but it really competes against cellular data networks, which are closed off, walled up, and expensive. There's no real competition among cellular providers, each of which offers similar rate plans ($60 per month for voice subscribers with other conditions; $80 per month otherwise), and generally similar network limitations.
Blomquist says that cities encouraging the building of Wi-Fi networks and paying for services as anchor tenants--that is, changing the payee on a line item for telecom services--is the equivalent of granting a monopoly. It's always remarkable to me that large entrenched businesses are never counted as monopolies, even when they exhibit all the tendencies of such, and that freeing them from the shackles they bear (made of paper clips at this point, perhaps) will cause dramatic increases in services and a drop in prices.
Yeah, right. Let's close this wormhole to 2005, and talk about the realities of 2007 again.
Step right up, step right up, and win a wireless cupie doll! NetworkWorld files this amusing and technically interesting report about the use of wireless networks at the California State Fair. The IT director for the carnival firm that runs this event comes in with a staff and lashes Firetide Networks 5 GHz wireless routers up wherever they can make a good line of sight connection to form the network's backbone. They even erect a 65-foot tower as the foundation of the network, on which they stick four directional nodes plus a long-haul link to another area. Firetide's main units mesh over 5 GHz with Ethernet ports to add LAN elements. Cables run down into ticket booths were access points and Ethernet-driven point-of-sale systems are run. Mobile scanners are also used to manage tickets, purchases, logistics, and personnel shift changes.
EarthLink tells St. Petersburg to pay, wait: The network provider won the bid for a 60 sq mi network earlier this year by "touting its cash on hand, existing subscriber base and contracts to build networks in larger cities," reports the St. Petersburg (Flor.) Times. However, the city's CTO said EarthLink has asked the municipality to wait until fall, when they'll "have their game plan together." That game plan includes a commitment to fees from the city, which the CTO said isn't of interest until the network is running. So that's pretty much that.
A local firm, Citi Wifi, has been building smaller networks, and lost out in the bidding. The firm would require the city to commit to paying to use the network before building it now, of course, which is only sensible.
Cleveland asked for free municipal service, got five bids, but we'll see where that goes: The Ohio city, notes MuniWireless.com, bid out for an operator that would give municipal workers no-cost access. Nonetheless, EarthLink, MetroFi, Azulstar, and two other firms responded. We don't know yet what they said, and EarthLink would likely bow out in any case. MetroFi almost certainly responded that they'd be happy to build a network in which there's no free service for municipal workers. Esme Vos notes that two firms with intent to bid, ARINC and CONXX, did not. The former (a giant) has done a number of transport-Fi network, and is part of one of the in-flight broadband/voice providers that might launch an offering soon. CONXX runs the network for Allegany County, Maryland, and is the only sizeable installation of Wavion nodes that I'm aware of.
EarthLink ain't talking, progressing, in Houston deal: The Houston Chronicle reports that EarthLink hasn't yet signed a contract with the local utility, despite the expectation that that part of the deal to build the 600 sq mi network would have been done in April. The utility forwarded a final contract a month ago; no response. EarthLink isn't providing comments at the moment on anything to do with muni-Fi.
In Houston, the company got the city to agree to $2.5m in anchor tenancy on a proposed $50m deal. EarthLink would also have to pay $5m in penalties to exit the deal. The clock on timing doesn't start until they sign the utility contract, which was clever on their part. They're not technically late yet.
As I'm quote in this article, it seems entirely unlikely that EarthLink will build Houston. They'll take the hit and move on. $5m is a lot to pay out, but if you're trying to shed short-term expenses, $5m versus the $20m to $30m in equipment costs is a small price to pay. EarthLink might be able to negotiate an exit strategy, too, of course.
Just an FYI for those who think that my and others coverage of problems in the municipal build-out means that metro-scale networks aren't going to be built: On the contrary. I expect that while networks in cities of 500,000 or greater will find fewer bidders and less interest due to political pressures and the difficulty in moving money from one budget line item to another, medium-sized cities and small towns will continue to build networks at perhaps just a slightly slower pace as the service provider market adjusts itself to the new realities. We'll see ever more networks that are dedicated initially to specific purposes. Once those purposes' goals are met--such as providing live video feeds for public safety and remote connections for municipal workers in a few departments, say--then the cities and their network partners will expand into other arenas. Public access may not be the first cited purpose of future networks, or at least not residential access.
I see a short-term dramatic decrease of interest in the consumer user and consumer market. It's pretty clear that reaching people in their homes is not the best or easiest first route to finding revenue. I've written for many, many months that if you build a network thinking that consumer, residential revenue will drive the network to break even, you're smoking Cat5 cable insulation. All that's happened in 2007 is that service providers have stopped hemorrhaging cash, and are re-establishing the focus.
Wi-Fi works very well as an outside coverage standard with current generation hardware of all kinds. This covers the vast majority of municipal and public safety purposes. Broadband wireless, as a broad category, works very well as wired T-1 and DSL replacement service. (That's Canopy, pre-WiMax, Proxim's MP gear, "real" WiMax in 5.8 GHz, etc., along with all the metro-scale vendors point-to-multipoint stuff.)
That is outside to inside doesn't work very well right now. What works? Outside to outside. (Also, Wi-Fi works quite well as an inside to inside technology.)
What's vital to understanding my viewpoint that Wi-Fi networks will be built over more and more cities, even if at a slower pace, is that gadgets, not residential usage, will drive consumers onto networks. The iPhone, T-Mobile's HotSpot@Home unlicensed mobile access, the huge number of cell phones and tablet PCs and handhelds with Wi-Fi built in, the coming tsunami of digital cameras with Wi-Fi built in, the coming-soon eyeFi camera adapter--this isn't a fad.
Just like Bluetooth, gadget Wi-Fi is multiplying without a lot of big picture attention. About Bluetooth, many people said the complexity, battery drain, and lack of consistent support would doom the standard. Then, seemingly all of a sudden, Bluetooth was in millions then hundreds of millions of devices. Utility outweighs the minor problems in getting it to work. (This is the inverse of Douglas Adams' description of products made by the Sirius Cybernetics Corporation in Hitchhiker's Guide to the Galaxy: the products' "fundamental design flaws are completely hidden by their superficial design flaws.")
Now, while Wi-Fi is found in tens of millions of devices, laptops, and desktop computers, it's still incorrectly seen as a host-based, computer-oriented protocol. New crazy stuff like Nabaztag and Chumby should put the lie to that, as well as the tidal wave of products in the pipeline that we'll see at Christmas and in 2008. Every kind of device is going to have Wi-Fi. And every consumer will be walking around with one or more Wi-Fi devices needing access. And they'll be outdoors. And they'll be happy to pay a few dollars a month per device.
That's my vision of the way the market shapes out. Tune in to this Wi-Fi channel to see how my prognostication pans out.
A digression on indoor bridges: To reach indoors is an expensive problem for consumers and service providers. Consumers have to buy $100 to $300 Wi-Fi bridges or service providers have to subsidize that cost. I recall back in 2005, talking to Tropos and asking what kind of customer premises equipment (CPE) they sold. They said, we don't offer that gear. We let the service provider or city figure that out. We recommend high-gain Senao gateways. I was baffled. I thought I must be missing something. In fact, it wasn't me, it was the whole industry.
Now, a sub-section of the industry has evolved to sell bridges, but those bridges are still expensive, because higher-gain radios (200 milliwatt/mW and 400 mW radios) cost substantially more than 30 mW and 100 mW radios more typically found. The folks at Meraki Networks say they spend about $5 in parts for their $49 Indoor units. Tropos et al don't reveal their raw-cost prices, but they're all using Atheros chips. It's more likely that for a 2.4 GHz radio of the up-to-the-legal-limit power that all the metro-scale vendors offer, they're spending $200 to $400 in raw parts before assembly. (hField just released their latest Wi-Fire model, a 500 mW EIRP USB adapter/antenna combo for Mac and Windows that could be an interesting option for indoor laptop users; I haven't tested it.)
Recently, Phil Belanger of Novarum gave me some good news. His firm tests metro-scale networks using a variety of clients and hardware, selling reports to service providers and cities, and providing independent evaluations. Belanger said that after adding an 802.11n adapter to their test bed, they were very surprised at how usable some networks suddenly became. "The consumer clients, the crappiest pre-N consumer clients, work really well on a metro-Wi-Fi network. In fact, as well as or better than a high-powered PepLink or Ruckus device. That was astonishing to us," he said. (Of course, Ruckus and PepLink are certainly working on 802.11n products, too; both companies' current range of adapters use higher power to reach further, although Ruckus also beamforms with multiple antennas.)
Which means there's some relief coming for service providers who are wrestling with the CPE issue. Instead of wrestling with bridges, their cost, and the messaging to consumers, they can rely on a new generation of 802.11n that's already shipping with equipment from Apple, Dell, and others, and that Belkin and other firms are making available as USB, ExpressCard, and PC Card add-ons.
It's not a guarantee of coverage, but it certainly increases the odds of a good connection, reduces cost, and eliminates most of the wonkiness and complexity.
Forgive me for taking a week to read the FCC's order governing the licenses for the 700 MHz band they'll auction off a few months hence: The FCC put out its press release two weeks ago about how the terms by which licenses would be auctioned and usage of them regulated for a bunch of "beach-front" spectrum--frequencies ideally suited to reach into homes with less power and less equipment than needed by most currently allotted bands. The most significant of the licenses are a set of six (the Upper 700 C Block in the auction terms) that span the country and offer now 22 MHz (split into up and down pieces) that could deliver 50 Mbps per cell based on the current roadmap for cell and related wireless standards.
The commission just released its Order and Report last Friday, however, which weighs in at 312 pages, 1204 footnotes (excluding commissioners' statements) and 567 numbered items in the main body. My interest weighs largely in the area of how the commission defined the form of open devices and open applications that were requested by Google, Skype, other Internet firms, and a number of public interest groups, among other parties. Google's desire to have a non-discriminatory resale requirement for the C Block that would allow any provider to offer access for sale, paying reasonable wholesale rates. That wasn't agreed to. But open applications and open devices could still change the market.
My concern was partly due to the press release's loose language, which conflated handsets and devices. An open handset requirement might mean you could use any phone you wanted on a 700 MHz network provider's network. (It's most likely a single provider will buy all six licenses, providing national coverage.) But an open device requirement means that the plethora of equipment available for Wi-Fi could find its way in some form to 700 MHz C Block. Right now, getting permission and certification for any device to use a cell network requires a lot of time and money. The folks at SmartSynch, makers of wireless utility meters, said it costs them $150,000 to get cellular certification. The same device in Wi-Fi form costs $10,000. The difference are all the layers of additional approval, at the end of which, the carrier still has to agree to allow your device on the network.
The discussion in the report and order start on page 75 (that's III. A. 2. a. (iii), item 189, to be precise). For a Republican-controlled FCC, the report and order sounds awfully jaded. The commissioners and staff have apparently experienced enough nonsense from carriers control of their networks first hand to sound a little pissy. In item 198, for instance, "Although wireless broadband services have great promise, we have become increasingly concerned that certain practices in the wireless industry may constrain consumer access to wireless broadband networks and limit the services and functionalities provided to consumers by these networks." Fair enough, but then: "wireless handsets with Wi-Fi capabilities have been largely unavailable in the United States for reasons that appear unrelated to reasonable network management or technological necessity."
Further, the FCC notes that while competition among carriers is robust, covert behavior with imperfect knowledge leads to an imbalance: "while it is easy for consumers to differentiate among providers by price, most consumers are unaware when carriers block or degrade applications and of the implications of such actions, thus making it difficult for providers to differentiate themselves on this score." Hey, they're starting to sound mad. "there is evidence that wireless service providers nevertheless block or degrade consumer-chosen hardware and applications without an appropriate justification."
The FCC also wants equipment makers--there's the business interest--to have more ability to sell their goods. "By fostering greater balance between device manufacturers and wireless service providers in this respect, we intend to spur the development of innovative products and services." Very Wi-Fi of them; Wi-Fi is mentioned several times as a combination of light regulatory touch and successful innovation. They also note (item 205) that even though the rules would apply just to a single 700 MHZ band license, that the experiment, if successful, would promote operators to adopt these rules in other bands.
In 206, the FCC defines what it means: "Accordingly, consistent with the broadband principles set out above, we will require only C Block licensees to allow customers, device manufacturers, third-party application developers, and others to use or develop the devices and applications of their choosing in C Block networks, so long as they meet all applicable regulatory requirements and comply with reasonable conditions related to management of the wireless network (i.e., do not cause harm to the network). Specifically, a C Block licensee may not block, degrade, or interfere with the ability of end users to download and utilize applications of their choosing on the licensee's C Block network, subject to reasonable network management. We anticipate that wireless service providers will address this requirement by developing reasonable standards, including through participation in standards setting organizations, as discussed below."
After a long discussion of the FCC's right to make these rules--interesting reading that covers the first amendment, commercial speech, and other matters--the FCC gets into particulars in item 222. Wireless providers won't be able to turn off features on handsets. Meaning, if they sell you a phone and it has Wi-Fi built in, the Wi-Fi has to work. They also note a list of things that must be allowed: "We also prohibit standards that block Wi-Fi access, MP3 playback ringtone capability, or other services that compete with wireless service providers' own offerings. Standards for third-party applications or devices that are more stringent than those used by the provider itself would likewise be prohibited."
In terms of bandwidth, the FCC offers some key protections: "In addition, C Block licensees cannot exclude applications or devices solely on the basis that such applications or devices would unreasonably increase bandwidth demands. We anticipate that demand can be adequately managed through feasible facility improvements or technology-neutral capacity pricing that does not discriminate against subscribers using third-party devices or applications."
In short, carriers can't define harm arbitrarily, and the FCC elsewhere describes enforcement action for behavior contrary to its rules. This means that, for instance, if Verizon wants to charge a flat rate for downloads from its own music service but charge by the megabyte at a disproportionate rate for all other downloads, they would likely be in violation. Better, the companies whose services would be affected would most likely be multi-billion-dollar firms who would pursue immediate injunctions. On the last go round of regulatory action against incumbent violators of access rules, the damaged firms were startups struggling with technology and limited budgets. Now we have Google, Intel, Apple, Real Networks, the record labels, the film studios, and many others.
In item 223, the commission says that providers can still use their same certification standards, which could keep costs high. But the processes have to be more limited than those that are currently in place, as the process must be limited to just "reasonable network management" standards being tested.
So that's that. The costs will be higher to get 700 MHz devices made and certified, but almost certainly not as high as for other cellular bands. Further, the operator gets to control the process, and if Google were (unlikely as it seems now) to win the auction for the C Block, they could choose to have extremely minimal certification processes beyond the FCC's own requirements. What these rules do ensure is that a device maker has predictable access to spectrum, and that customers can access services designed to work specifically with devices, and that pricing for such access has to be in line with what an operator charges customers for its own comparable services.
Now we just wait for the auction, which must occur no later than January 2008. The spectrum itself will likely not be unencumbered until as long as the digital television transition deadline of Feb. 17, 2009, when analog TV dies.
The Long Island network deal raises a few eyebrows: The New York Times sticks "newcomer" right in the headline of their story about Suffolk County and Nassau County's selection of e-Path Communications to build the largest currently proposed single-contract network. The counties chose e-Path and its partners Cisco and KeySpan because the municipal entities didn't have to cross the consortium's hands with filthy lucre. Here's the money graf from the article:
"Some Wi-Fi systems elsewhere have fallen short in their expected advertising revenue, and [Suffolk County executive Steve] Levy acknowledged that it remains to be seen if e-Path’s innovative financial plan will succeed. 'Let's see how this works,' he said." Yeah. That's how I plan a reportedly $150m, 700 sq mi network, too.
(Wireless Silicon Valley's 1,500 sq mi requires 40-odd individual contracts with cities and counties; Houston's contract with EarthLink covers 600 sq mi. A network in Eastern Oregon is larger and growing, but covers a population that numbers a fraction of Long Island's two counties involved in this deal.)
Newsday follows up on their brief account yesterday of the deal by surveying how other networks are faring, and whether the model proposed by e-Path et al can work. I'm quoted saying, no, it can't. More significantly, the reporter spoke with Chuck Haas, CEO of MetroFi, which was one of the bidders on the Long Island network. "That company bid on the Long Island project, including upfront costs in its plan, an expense county executives sought to avoid. But Haas said lack of upfront money makes the business model unsustainable."
It's critical to recall, too, that "upfront money" should really be described as "fees currently paid to a variety of telecom and data providers, mostly incumbent monopolies, that would be transferred to a new wireless operator as the network achieves benchmarks of performance." There is plenty of money to be moved over or saved from city telecom budgets for dedicated wired lines, among other areas, where the service commitment to a MetroFi or EarthLink would involve no incremental increase in expense, or perhaps a one-time increase followed by a decline.
Joining in on the fun of describing the change in the muni-Fi market, the Wall Street Journal provides its take: This article details the cost overruns, delays, and business model problems with many networks that are being and are not being built. Note that everyone quoted is now in agreement that cities need to step up to the plate and agree to purchase services to have their networks built. That used to be a narrowly held opinion as recently as early this year. As reality has set in, so, too, has the attitude changed. One note about the article, though: It repeats the common fallacy that "Initially, cities funded their projects out of their own budgets." Only certain smaller cities chose that funding route. Some larger towns planned to fund their networks that way. But I know of no deployment of any scale that didn't involve a vendor bearing all the cost. Networks in Chaska, Minn., Lompoc, Calif., and St. Cloud, Flor., were built at city expense; the former two as for-fee networks, and the latter as a free one. Corpus Christi, Tex., built the largest city network I'm aware of, but for municipal purposes, selling it to EarthLink when the transition to a for-fee public access network came into view.
What about the equipment vendors? It's been a while since I spoke to Tropos, BelAir, Strix, Motorola, Cisco, and SkyPilot about their metro-scale gear, because the focus of the industry (and this site along with it) moved to deployments and applications rather than the hardware. How does your business model change, when networks that would require tens of thousands of nodes are put on hold? I expect we'll see the answer to that question by year's end. One part of the answer is that the four startups in the list above may have promoted themselves as municipal solutions, but are selling products worldwide and for corporate campuses and private purposes. I expect we'll also find none of these firms were silly enough to put all their eggs in one basket.
Olga Kharif writes in BusinessWeek about the flat-fish nature of the current municipal wireless market: Kharif plays the kazoo--as I have been--at the pre-wake for the idea of freely built city wireless network that may or may not offer free access to residents and visitors. Freely built networks involve a firm agreeing to bear all costs, with no commitment of revenue from the city, businesses, or residents. Those are dead.
The new model involves upfront service contracts from cities that replace existing telecom line items, ostensibly offering cost conservation or reduction in fees. But it's unclear how rocky the transition will be between the current model and the new one, with so many networks in rough shape during construction, and EarthLink ostensibly working with existing contracted cities to figure out how to make the business model work.
I'm quoted in the article noting how my research shows that where numbers of subscribers are reported, the level of subscriber is always enormously below the long-term break-even point, even on networks that have been operating in some fashion for a year or more. I also found that metro-scale service providers stop giving out numbers after initial excitement, which can only be attributed to a failure for those numbers to improve. While few networks beyond smaller towns are fully built out, one would expect that in areas covered, subscriber rates would climb towards predicted penetration levels if residents were able to use the service in their home. This is another data point in my book for how poorly in-home coverage from outdoor Wi-Fi nodes works.
Kharif mentions Lompoc, Calif., which appears to be the poster child for how these networks wind up not meeting expectations. The Lompoc Record said a few months ago that the city had spent $3m so far. Kharif reports that with 40,000 residents, they have just 442 users, hoping to grow to 1,000 by the end of the year. That's with a price cut implemented a few months ago. The original break-even point was 4,000 paying subscribers.
Meraki builds network from small pieces, loosely joined: Meraki has 6,500 regular users of its Free the Net network, which has grown to cover a hunk of Mission, Haight, and Alamo squares using about 200 nodes, according to the company. They had just 50 nodes and 1,500 users a few months ago; growth is rapid. They have a live map showing overall users, active nodes, and bandwidth transferred. Meraki's mesh networks allow multiple entry points for backhaul bandwidth, and communicate with each other using a slightly modified form of Wi-Fi; end users can use normal Wi-Fi to connect. At $49 a node for the indoor nodes, it's pretty cheap to expand the network. Meraki has 1,500 volunteers involved in the network now, and believes 15,000 will get involved by the end of the year. Strange how bottom-up networks seem to be building where top-down is proving more complicated, expensive, and elusive.
Long Island's Suffolk and Nassau counties pick new firm for unwiring: e-Path Communications got the nod in a field that included an IBM consortium, MetroFi, and EarthLink. (I don't know that the latter two firms stayed in the bidding at the end, since bids were solicited early in the year.) e-Path Communications, a Florida firm (which apparently likes the "e" logo from Microsoft Internet Explorer; see their site), was founded in 2006, and will work with Cisco and KeySpan, a Northeast natural gas utility that offers service in those counties. KeySpan also owns 80,000 miles of fiber optic lines. 2.7m are covered in the two counties across 750 sq mi. Cisco's involvement probably involves financing, as they have elsewhere. The network's cost to build is estimated at $150m.
Newsday reports that this is an "old school" deal, with the consortium not requiring any fees from the counties. Limited access will be available for free outdoors--which means for mobile use without a booster, ostensibly--while higher tiers of service intended for homes will start at $25 per month and businesses at $55 per month. ePath is currently working in a joint venture with ATCI (a firm that mostly deals with high-tech security) to unwire Delray, Flor. It's hard to imagine how a deal like this is financially feasible without service contract commitments from the county.
Minneapolis and St. Louis Park don't know how good they got it: A short piece by the marvelous Steve Alexander, a business and technology report for the Minneapolis Star Tribune about the "bumpy" road that the two cities have had in getting Wi-Fi networks up and running. The delays they've seen are so miniscule compared to those in other cities, they should consider it smooth sailing. Minneapolis took forever to pick a vendor, and US Internet has a few technical hurdles to deal with.
The operation of Britain's east coast mainline will switch in December; new provider promises free Wi-Fi for all: In the UK, rail infrastructure and operations are uncoupled, and the government awards contracts for lines to private firms. The east coast line has been run by GNER for several years in a generally positively received manner, but the firm can't afford the franchise fees going forward. National Express will take over in December of this year until spring 2015, paying nearly £1.4b for the privilege.
GNER was the second major rail line in the world to offer comprehensive Internet access; SJ was the first. Both projects originated with Icomera. GNER uses cellular backhaul for service, and the growth in 3G networks and the increase in speed allowed them to achieve quite comprehensive coverage and reasonable speeds. GNER charged for access in coach, providing it free to first class. National Express will offer free service in all classes.
Competition for the line was heated, and National Express kicked in free Wi-Fi among other incentives to get the bid. They're also promising to refund ticket fees if a passenger has to stand for an entire journey.
News outlets are reporting on the FCC filing by T-Mobile for a VoIP/Wi-Fi router: The filing shows an unannounced Linksys WRTU54G Wi-Fi router, much like the ones that T-Mobile is selling along with its HotSpot@Home service to ensure the best quality of service and battery performance for their Wi-Fi/cell handsets when used in the home. The only difference? The router up for certification has two phone jack plugs, similar to those found on telephone adapters used for VoIP services, including other models sold by Linksys.
The fact is that unlicensed mobile access (UMA), the technology underlying T-Mobile's converged calling plan, matches regular GSM calling over cellular networks with VoIP over the Internet via Wi-Fi. The VoIP part is encrypted using GSM technologies, but between the Wi-Fi/cell handset and the Internet portal on T-Mobile's network where the voice conversation pops out, the call has all the problems and benefits of pure VoIP.
Writers at TG Daily (the original source of this information) and News.com mistake the reason for the router's ability to accept up to two GSM SIM cards. The writers talk about how it might be "merely a way to get up to two phone numbers into the WRTU54G" (TG Daily) and that the cards "would also allow users to add up to two additional cell phone lines" (News.com).
Not right. To perform GSM authentication and encryption, a device has to have a SIM. For the router to work interoperably with T-Mobile's UMA gateways, it has to make VoIP look like encapsulated GSM, which means that a SIM is required. A UMA handset treats a Wi-Fi network like another cell tower. A router with landline-style phones that can make UMA VoIP calls only on the Internet side is actually three layers of pretense: A landline phone is pretending to work like a landline phone; the router is pretending it's on a GSM network; calls placed are pretending that they're being made from a cell handset.
It's a little contrived, but it allows T-Mobile to leverage 100-percent of its existing infrastructure, with perhaps a slight increase in cost on the SIM side. The routers don't add any additional cost to T-Mobile handling calls, except increased call volume, but 100 percent of that volume would come on the Internet side, far cheaper to handle than on the cellular side.
The big issue in my book is pricing. T-Mobile's requires at least a $40/voice subscription to use HotSpot@Home. You can then pay $10/month for one line or $20/month for two to five lines for unlimited Wi-Fi minutes, or choose to use minutes from your cell pool for Wi-Fi calls. (The rate rises this fall, apparently, to $20 for one line and $30 for multiple lines, but the initial rate applies indefinitely to anyone who signs up in the early period.)
Would T-Mobile decide to include the one or two lines in the Linksys router as FamilyTalk additional lines, lumping them into the monthly multiple-line fee, in the interests of making sure to capture more revenue, and perhaps convert more family members on the cell side? Or would there be an additional fee, perhaps $10 per month, to acknowledge the additional calling that would take place? Would integrated voicemail across multiple lines be provided? Could you easily forward your cell to the landlines so that as you arrive home, your calls come in on a cordless not a cell? Is this another tool in T-Mobile's arsenal against Vonage et al.?
A lot of questions remain to be answered, but it could be a unique combination of services that would increase ARPU (average revenue per user), especially in families, while decreasing the cost of delivering service, and decreasing churn.
Someone stuck a stick in the sand and said it's time to declare the network finished: This report from the local paper notes that the network isn't done, but they "celebrate the completion" anyway. It's unclear what state the network is in beyond still in progress as EarthLink densifies the $5.5m network they purchased from the city. They'll pay Corpus Christi $340,000 in the first year, and other fees in future years.
The article fail to mention that this could be one of the last celebrations of this kind that EarthLink participates in. EarthLink has re-evaluated its metro-scale network business model, and only cities underway remain on track. I have said multiple times, based on the situation in San Francisco, that it's likely EarthLink withdraws from that proposal now in a matter of weeks (months ago, I said months); I expect that the Houston deal for 600 sq mi will also be scotched.
Milwaukee, Wisc.'s long-delayed Wi-Fi network faces a further elongation of the schedule to as late as "never": Midwest Fiber Networks agreed to build the city's network as part of a deal that would allow them to leverage their fiber infrastructure and services. I don't know the Wi-Fi equipment vendor they chose--last I can find, they were evaluating equipment 18 months ago from several companies, including Tropos and Cisco--but the network just isn't robust enough in their testing. (The picture shown with this linked article is clearly a Tropos node, but I'm not sure if it's a file photo or not. Update: A reader says that Milwaukee went with Cisco gear.)
The usual suspects emerge: Inconsistent performance, lack of coverage, more nodes will be needed, etc. Midwest Fiber might abandon the plan altogether. They are studying "whether it will make financial sense to build the network." Sound familiar?
At the launch of Virgin America's service--yes, their virgin flight--they openly discussed having AirCell in-flight broadband next year: This, according to Xeni Jardin of BoingBoing, who reported from their launch. She confirmed via email that AirCell was being openly discussed by Virgin America as its Internet supplier. She noted in her BB report that the planes have two 110-volt outlets for each three seats in coach, one per seat in first class; Wi-Fi access points in the front and rear; and Ethernet at every seat. There's onboard chat services, too, to allow intraseat communication. There will also be pay-per-download music sales. They'll also deliver other Internet content onboard, presumably via a caching media server or proxy server.
Jardin also notes that the plane is, you know, comfortable. Lighting isn't harsh: "The cabin was softly lit on our daytime flight in purple and pink, the mood lighting is different at night." Their initial routes are between SFO and LAX, JFK, Dulles, and Las Vegas; JFK and LAX; and LAX and Dulles. (Can we beg them to enter the Seattle market for nonstop cross-country?)
Contacted for comment, AirCell said that they are in discussions with several airlines, that they neither confirm nor deny arrangements with any non-announced airline, and that they are covered by non-disclosure agreements with the companies with which they are negotiating, and thus can't provide any detail until such point as a contract is final.
Apple slips in 1000 Mbps Ethernet in its Wi-Fi router: Apple quietly upgraded its Draft N-based AirPort Extreme Base Station to full gigabit Ethernet support across its three LAN and one WAN ports today. While the company didn't release information separately, they contacted me to note the change, and the Apple Store's product listing has been updated. The new base station can be ordered right now.
The first release of Apple's Draft N base station was rather marvelous for its inclusion of a USB port to share multiple printers and hard drives; the company's decision to have both 2.4 GHz and 5 GHz radios inside; and the fact that Macs had been shipping with 802.11n inside, requiring just an enabler, released with the base station, to upgrade their performance. My primary complaint, however, was the mismatch between the company's widespread inclusion of GigE in most of its models long before the competition. With the drop in cost in GigE switches, it seemed odd for Apple to release a unit that was designed for homes and small offices that would underperform a $35 Ethernet switch.
I also suspected that the overall performance of the 802.11n draft that Apple is using was constricted due to internal Ethernet limits. In my testing for a review in Macworld, I was able to top 90 Mbps in Wi-Fi to Ethernet and Wi-Fi to Wi-Fi transfers. But Ethernet-to-Ethernet data was limited to just over 90 Mbps as well. Apple says that their new gigabit Ethernet base station is up to 50 percent faster for wireless-to-wired links, which would put it closer to 150 Mbps, a speed achieved on other GigE-based Draft N routers.
When testing the base station in February, I discovered that with NAT enabled to share access from an incoming WAN link, performance was restricted to about 30 Mbps from wireles LAN to LAN and 60 Mbps from wired LAN to WAN. Apple confirmed this was a bug that was due to performance issues in their NAT stack. Apple wasn't able to tell me if this limitation has been fixed, but I would imagine so.
This bug emerges in only two edge cases: Where a broadband connection exceeds 30 Mbps, which is true for some fiber and cable customers; or where a corporate or office LAN isn't supplying addresses to the computers connected via the AirPort Extreme. If NAT is turned off, the AirPort gateway has no performance limitations.
The price for the AirPort Extreme Base Station with Draft N remains $179.
The New York Times agrees with me: The Times unsigned editorial says that the FCC's failure to adopt a resale requirement for the 700 MHz nationwide unencumbered commercial licenses (22 MHz of prime territory) wasa big failure: "American consumers have once again been denied a truly open and competitive cellular market."
SF Bay Guardian editorial says buh-bye to EarthLink: The local weekly writes in an unsigned editorial that there's no likelihood of EarthLink agreeing to SF's terms, and no chance that SF would "[fork] over millions of dollars a year to guarantee EarthLink some baseline revenue." The Guardian says good riddance, which has been their attitude all along, as they'd prefer a comprehensive fiber-optic plan.
White space device fails FCC trial: Google, Microsoft, and other firms submitted a device to the FCC for testing that would allow low-level signal use of "white space," or unused broadcast spectrum which varies depending on the TV market. Unfortunately, the FCC found that the prototypes didn't detect TV signals accurately enough to prevent transmissions when broadcasts were present. The FCC didn't close the door, noting that the prototypes were "initial efforts," according to IDG News Service.
Clearwire increases customers, revenues, losses: The wireless firm has 299,000 customers, up 41,000 in the last quarter; $35.5m in revenue, a jump of 32.4 percent year over year; and a loss of $118m. Their head notes that in established markets, 14 of 25 areas served are cash-flow positive. They also have over 10 percent of households subscribed in those markets. Clearwire said it will start testing 2 Mbps down with 4 Mbps burst for $45 per month for residential users; and will test a PC Card with a $60 per month subscription.
The Bush Administration declined to use its veto to overturn a trade ruling that will prohibit the import of cell phones and devices that use Qualcomm third-generation (3G) chipsets: The ruling, in which Broadcom's claims of patent violation were found to have merit, prohibits the importation of any model of device that wasn't already being imported before June 7.
Verizon sidestepped the matter by agreeing on a fee schedule with Broadcom, that included Verizon withdrawing its support for Qualcomm's lobbying and legal efforts. Verizon will pay Broadcom $6 per phone with the infringing chips, up to $40m per quarter and up to $200m overall. A drop in the bucket if they have advanced phones that their competitors don't, such as the new Blackberry with Wi-Fi that AT&T planned to introduce this month. I do not know if that Blackberry model would be covered under the ban, but it's possible. (GSM 3G chips have a number of suppliers, as opposed to Qualcomm's control of CDMA, a standard they invented.)
Qualcomm has another tool, however; it immediately announced that it would ask a federal court for an injunction now that the US Trade Representative, to whom Pres. Bush had delegated this particular veto power, has opted against intervening. A previous attempt at an injunction was turned down by the Federal District Court on the grounds that it lacked jurisdiction until that decision had been made.
Minneapolis's Wi-Fi provider made three great choices in the wake of the bridge disaster, Julio Ojeda-Zapata writes in the (St. Paul) Pioneer Press: The made it free, expanded it, and added full zoom/pan/tilt Web cameras.
Meraki gets a long, positive write-up in Scientific American: The magazine does a great job of hitting the high spots about why Meraki's idea of cheap, mesh routers that self-organize and can take multiple injection points without configuration have had such positive early impact. Most of their competitors are selling devices for 40 to 100 times as much as their $50 box, but those more expensive devices are designed to be more rugged, often work with public safety networks, feature radios with 10 to 35 times the raw signal power, and so forth. (Radios grow more expensive on a steeply climbing curve as you increase power output.) Meraki's raw cost? About $5 in components (chips and the radio); the rest is circuit boards, assembly, cases, and overhead.
The Mercury News notes that the Wireless Silicon Valley plan now has an anchor tenancy requirement of sorts: Another shoe drops. The Wireless Silicon Valley project was supposed to offer free access at a reasonable speed to all residents covered by the network, ultimately, and it was to be funded by the consortium building it, which includes Cisco and IBM. The free requirement is still in place, and the network builders are still putting out the initial sums. However, the reporter notes that there's been a rather significant shift: the cities covered "will be asked to pay to use the network, enabling police officers, firefighters and even housing inspectors to file reports on the go." That's no secret: It was clear from the beginning that there was the intent for cities to sign up. But it's clearly moved from the consortium encouraging cities to commit to contracts at some point to requiring cities to commit before the network is built. Which, by the way, is vastly behind schedule, and no model contract has yet been publicly seen.
I wrote the first of these stories two years ago about cafes turning off Wi-Fi or changing their model after being deluged with barely- or non-paying laptop table campers: The latest installment looks at a few places in New York, where store owners are a little more than frustrated. Waltz-Astoria is charging $2 per hour for Wi-Fi or electricity, offering it free to regulars who sign up for their email newsletter. The Roebling Tea Room and another shop, Aroma, have covered their electrical outlets. Aroma shuts down Wi-Fi during lunch.
These campers spend from $0 to $3, stay as long as eight hours, and occupy tables for four with their lonesome, and use a few pennies or more of electricity, to boot! Now, they're a subset of all cafe laptop users. Myself, I'm often in some venue for an hour or two, spending $3 to $5, and taking the smallest possible footprint. It only takes a few bad apples, though.
My early story--tipped by a colleague who spotted the trend--covered Victrola here in Seattle. My editor back then asked, Is this a trend? Or is it just a few places? I said that I was pretty sure it was not a trend, but it was an interesting development that was likely to spread slightly as cafes, looking for an option, heard about the idea. Once you install Wi-Fi, it's weird to turn it off or change the terms of access; with a peer group of other owners, even those you don't know, it may be easier to modify what you're offering. [Link via Craig Plunkett]
Ricochet has gone through more lives than a cat stuck in a rotary press (I'm no Dan Rather): The first city-wide wireless networking system has changed hands yet again, the fourth time in six years, and shut down its San Diego operations in favor of focusing in Denver, where it has 6,000 subscribers. Ricochet's latest move announced today is that the company's parent has sold it to Civitas Wireless, a company founded by Ricochet Wireless's president (since 2005), Judi Evans. She is the president, CEO, and majority owner of the new firm, the press release notes. Ricochet scored a deal with Denver in April that allow it to use its existing right of way to deploy Wi-Fi, WiMax, and other wireless technology. While Ricochet spins this as a deal with the city and county, the government never made such an announcement; the deal was really about using real estate, not about a public/private partnership.
Way, way, way back in 1997, when Paul Allen invested in this little idea about putting high-speed, mobile-enabled wireless networks using unlicensed spectrum across entire cities and in airports, the only problem was that high-speed meant 20 to 30 Kbps (later lifted to 128 Kbps). The unlicensed spectrum in question was 900 MHz, which propagates further, but the Ricochet technology was fairly inefficient and noisy compared to 802.11b, which emerged shortly after Ricochet's launch. The service did work, and was a lifeline for people who needed mobile access and would have spent another $20 to $40 per month for a dedicated phone line at home if they couldn't get broadband. (Read this great user review from about 1998; another by Joel Spolsky of Joel on Software from 2000.)
While Ricochet was mobile (external, battery-powered modem), it was expensive ($80 per month is the rate I'm finding), and started coming into its own just as broadband was hitting double-digit percentages of people's homes, and "56K" modems were becoming popular. If you could get 1 Mbps or faster at home over broadband or even 56K over a $20 per month dial-up service, you didn't need Ricochet. Further, cellular operators had already integrated modem-speed networking into cell phones and PC Cards, offering a taste of mobile networking that was often easier and more ubiquitous than Ricochet.
Metricom was fairly unwieldy. They raised hundreds of millions--including $300m from Allen's Vulcan Ventures and $300m from the late, unlamented WorldCom--and built the network fast. But the technology was out of date by the time it was installed, and the service price too high for casual users, and not useful enough for business travelers. Hot spots started to appear nationwide in 2001, and the combination of broadband, early cell data, Wi-Fi, and early hotspots just pithed Ricochet's potential.
Metricom ultimately filed for bankruptcy in 2001, and its assets were sold for about a penny on the dollar to Aerie Networks ($8.25m) late that year. In summer 2002, Aerie resurrected the network in Denver and later San Diego. Aerie sold Ricochet to EDL Holdings in 2003, which then sold it to YDI Wireless (now Terabeam) in 2004. (YDI took its new name from the acquisition of Terabeam, which itself once had over $500m poured into it, until it was sold to YDI for a tenth that. YDI's business plan has been to buy flagging wireless firms and then bolster their product lines. They also bought Proxim and KarlNet.)
For sheer chutzpah, you just have to love this item in The "About Civitas" section of the press release: "Civitas Wireless Solutions, LLC, d.b.a. Ricochet Networks, is the only Managed Service Provider in the wireless Internet industry with a decade of experience operating large-scale wireless deployments." Well, technically true, but puhhhhhhh-leeeeeeeze. That's only true if you count some of the equipment mounted in Denver as continuous employees of the company.
Network World Canada provides a little back and forth between telco Rogers Communications and Toronto Hydro: The former backs WiMax as the coming thing, so why bother with Wi-Fi. The latter once had plans for a large city-wide Wi-Fi network, and defends that technology's ubiquity and cost. Mobile WiMax isn't yet mobile; Wi-Fi is. WiMax in Canada is enormously more expensive in this pre-standard version: C$45 to C$65 per month; Toronto's Wi-Fi, C$29.
Only problem? Toronto's Wi-Fi network hasn't increased in area for some time, and the head of the company, quoted in this article extolling Wi-Fi, fails to mention its tiny size. It was rated very heavily by Novarum, a testing firm, partly because of the density of access points installed. Their six sq km network has under 4,000 users, and they went say how many below. But they won't build the network further until subscribers top that figure.
Gavin Newsom and the president of the city's board of supervisors stuck a free wireless resolution on the ballot five minutes before the deadline: The measure is nonbinding, but asks voters whether they support the notion of free wireless throughout the city. It also asks if they'd like free ice cream and a pony.
The existing contract with EarthLink was already unlikely to move forward due to EarthLink's business model changes. The emendations to the contract requested by the supervisors' head ensure that EarthLink will ultimately back out. I give it another four to six weeks before the whole deal is over.
Which means that SF has to return to the drawing board. Kite Networks is undercapitalized (per their disclosures on their recent spinoff) to take on a task like SF without outside financing. Cisco, IBM, and Intel are tied up with their Wireless Silicon Valley (Cisco/IBM) and Sacramento (Cisco/Intel) consortiums. MetroFi wants an upfront commitment of service revenue, which doesn't seem in the cards, and it appears rather busy with Portland, Ore., and several other cities. US Internet is tied up in Minneapolis. There's no other firms of scale involved in bidding on and winning contracts like this except Clearwire, which isn't building out Wi-Fi.
Minneapolis city, citizens leaned on Wi-Fi network for information, telecom after tragedy: This interesting story from Computerworld demonstrates the best aspects of currently deployed city-spanning Wi-Fi networks: outdoor access in emergencies. The operator, US Internet, couldn't reach the city, so opened the network for free for 24 hours. Nice move. Usage climbed from 1,000 users to 6,000 users. The city used the network to relay information from the field, including detailed maps and large files. US Internet was able to add nodes near the bridge the day after its collapse.
While US Internet's CEO thought people with Wi-Fi phones could switch from the overloaded cell network to his Wi-Fi network, it's unclear how many people have such phones. US carriers don't directly support any such models, except T-Mobile, which can't authenticate to any network that requires a Web page interaction.
BusinessWeek asked my thoughts about the FCC's choices over rules in an upcoming auction for wireless licenses: I wrote this essay out the of the frustration of hearing regulators' choosing a set of rules for one chunk of the 700 MHz spectrum called "regulation" when it didn't suit the party's interest and a "lack of regulation" when it did. Google et al proposed that we have a single broad enough national spectrum (in this case, six licenses that together cover the U.S.) that had a requirement for openness. Open access for any legal device, any legal purpose, and any reseller. Instead, we have a carrier-forged policy that will ensure that it's just business mostly as usual.
The dancing rabbits reference is shorthand for the vast array of devices we have in the Wi-Fi junk band that's supposedly unusable spectrum in the eyes of cell carriers. The Nabaztag is a very weird use of Wi-Fi, but it shows how innovation isn't restricted by mere sanity.
Errata Security's Robert Graham showed how easy it is to grab tokens from Web traffic sent in the clear over Wi-Fi to hijack a session in progress: Almost every site that offers account logins uses a token stored in a cookie or appended to every link on a page (Amazon's original, pre-cookie approach) that serves as proof the user logged in successfully recently. Often, tokens time out forcing you to log in again to get a new one. The tokens are sent in the clear, and if you know the browser cookie name for the token, you can grab it over an open Wi-Fi connection at a hotspot.
Graham demonstrated automated tools to accomplish what he's calling sidejacking at the Black Hat security conference. The Ferret and Hamster programs work together to grab and sidejack a connection. It's a rather neat idea; it was clear to many that these tokens were a risk, but it raises the risk profile when it's demonstrated how easy it is to turn that into something practical. Graham noted, the BBC reports, that many appropriately designed sites require you to re-enter your password to perform account changes, and this weakness wouldn't affect that level of security.
The point of this kind of sidejacking would be to create an easy process for a cracker at popular hotspots to insert malicious code into MySpace or other social media sites. It's pretty clear that you could automate a tool to scan for social logins, grab the token, make the connection to retrieve and post a revised page with a payload--all of which could happen in seconds.
There are three solutions to this problem:
Graham's business partner David Maynor gets a booby prize: The Pwnie Awards are a self-organized set of security raspberries presented at Black Hat but with no connection to the event. The judges, all security researchers, gave David Maynor a Pwnie for the Most Overhyped Bug, with a description of the events last year in which Maynor and his colleague Jon Ellch appeared to say and then deny that they had found exploitable vectors in Apple's native wireless drivers.
While Maynor noted in February that he had actually found such vulnerabilities and reported them to Apple, he said he would shortly release the code that he used, in order to show that the potential was there before Apple released patches, and that the patches corresponded to what he and Ellch found. Maynor has never released this code. I pinged him recently to ask if he was now ever planning to; no response.
The Pwnie judges note in a rather fair way, especially considering some of those judges' general attitudes towards Apple and its security responses, "In the end, the only public information about Maynor's Wi-Fi vulnerabilities are hype, denial, a media frenzy, and a patch that may or may not have been based on Maynor's findings."
And it gets more interesting: Graham praises Apple's iPhone patches in advance of Black Hat: Back to Graham. Graham praises Apple for its rapid response to a series of bugs found in the iPhone that had already been demonstrated to show exploits. In the cell world, Graham writes, "Apple [h]as set a record for responding to security problems quickly."
Short, personal rant about the vagaries of technology and my position relative to such: I decided to switch DSL providers recently, and chose my incumbent telecom Qwest, which can underprice competitive DSL carriers. I've given in to that notion, as I could have twice the speed for half the cost. The process goes rather well, from ordering online, getting a telephone plus DSL bundled price, and receiving a modem in the mail. Then it goes downhill. That's not unusual for me. I've discovered hitherto unknown bugs in many common pieces of hardware, including the latest Apple AirPort Base Station.
The Qwest modem, which combines DSL handling with a Wi-Fi gateway, seems to work perfectly at first. It's a 2Wire 2700 HG, which turns out to be in the field for about a month. It trains up fine, handles the PPPoA (PPP over ATM) login, and tests show a healthy set of rates close to the raw 3 Mbps/640 Kbps that the modem's reporting. (The service is "up to 7 Mbps," so 3 Mbps is the best it trains to at the moment.)
Things start to go south. The modem stops routing. Every time I connect to it from a browser, it reboots. I call tech support and get the single most knowledgeable tech guy I've ever spoken to. We go through a ton of troubleshooting, all reasonable stuff (not "sir, did you power up the unit?"), and it seems to be back in shape after a factory reset. Then a few minutes later, it all happens again. I call tech support again, and once again get a phenomenally good tech on the phone. He determines it must be a defective unit. We're thinking bad Flash memory or something. He gives me an RMA and has another one shipped out.
Two days later, the new router arrives, I plug it in, and after a little bit of happiness, same thing. Hello, tech support! Another great techie. After some remote testing, and his own ability to restart the router by simply accessing it over their internal network, he's thinking it's a firmware bug that hadn't been uncovered. They're sending me another router so we can make sure it's not two bum units (the behavior is slightly different betwen the two).
The third tech had some suspicion that Apple's Bonjour networking, a form of multicast DNS on a local network, might be involved, as they had a different problem that was equally mysterious with an Actiontec modem model that turned out to be caused by the Bonjour network name of a networked computer having characters in it that the modem couldn't parse.
The third modem showed up today, and I disabled the Wi-Fi gateway entirely. It's working fine with an external access point. I figure I'll get the use of the network over the weekend wait until Monday to see the modem fails when I punch in my regular settings.
In a Journal article about the pitfalls of using cell phones during flights--the social problems, not technical ones--the reporter slips in some interesting news at the end: Nearly a year ago, I wrote "Would You Fly in Chattering Class?" for the Economist about the perils and potential for in-flight voice and data calling. At that time, it seemed a few months before an airline would launch such a service. It's now still a few months away, but apparently getting closer. There are a lot of concerns that people placing calls during flights would completely irritate others. Or, as former OnAir CEO George Cooper--a delightful man to interview--told the Journal, "passengers typically indicated they wanted to be able to make cellphone calls during flights 'but didn't want to sit next to anybody using one.'"
In this Journal article, which is short on numbers despite many published surveys on the topic, the reporter notes at the end a few fascinating facts, not confirmed by any of the subjects. First, Southwest has budgeted $100m for in-flight email and Internet access. They're talking to vendors now, including satellite operators. I don't believe satellite-based communications can compete with AirCell's terrestrial network, but it's possible that the economics have shifted enough to make that happen.
The second bit of news is related to that issue: Row 44 is apparently working with Alaska Airlines to test satellite-based Internet access. In a conversation some weeks ago with Row 44, they gave some detail about their claims of supporting tens of megabits per second downstream from a Ku-band satellite, the same satellite type used by Connexion.
Row 44 says their lease terms are better and their technology ekes every bit from the connection. John Guidon, Row 44's CEO, told me in May that their antenna specification is "vastly different from anybody else's specification including Connexion's original specification." As such their system has "the desirable characteristics of high bandwidth without extraordinary cost."
Row 44 says that they'll be competitive with AirCell--even bettering them by offering worldwide service with a single set of gear. The next several months should be quite interesting as companies pile into the domestic, trans-Atlantic, and European markets.
EarthLink asked for more time to review changes proposed to its San Francisco contract by a city supervisor: The board of supervisors might vote for a revised contract, but EarthLink wants more time to consider changes. As noted last week, EarthLink is re-evaluating its entire muni business. I said some weeks ago that it was likely EarthLink will drop its San Francisco bid because the city is 97 percent unlikely to agree to be an anchor tenant. The original contract has explicit language that forestalls the possibility, and this Chronicle article indicates that's just not on the table for a proposal the board would approve.
The Bluetooth SIG's board of directors approved 2.1+EDR (Enhanced Data Rate): The revision of the short-range personal area networking standard will reduce power consumption and greatly ease pairing, the association between two Bluetooth devices. The changes can be applied to 2.0+EDR devices via firmware, the Bluetooth SIG told me, but many Bluetooth modules are now in devices that lack firmware upgrade ability. So computers, yes; phones, many; picture frames, perhaps not so much.
Pairing has been dramatically improved by reducing the number of steps and the complexity. For devices that require a passcode entry, version 2.1+EDR requires that one device in the paired set generates a six-digit PIN that is then entered in the other device. And you're done. (Apple created their own version of this years ago, but it worked only when devices were discoverable and paired by Mac OS X to a computer, and it was far less secure than the 2.1+EDR version.)
The PIN is generated, by the way, using an Elliptic Curve Diffie-Hellman algorithm, which avoids man-in-the-middle (MitM) attacks by using an out-of-band method to confirm a key exchange. In this case, the two Bluetooth 2.1+EDR devices generate and exchange their ECDH keys, and then one device generates a six-digit PIN which is part of a hash of the session key being used by the two devices. While an MitM can talk to both parties, they can't know that six-digit PIN.
The improvement in power usage is rather significant: the SIG reports a fivefold improvement in battery life by intermittently connected devices like sensors, and input devices that send very little actual information, like keyboards and mouses.
A well-received study that tested the effects of exposure from cell base station transmissions on self-identified sensitive individuals and control subjects has released more details to counter criticism: The University of Essex researchers reported that self-selected electromagnetically sensitive subjects were unable to determine when a base station signal was present or absent in double-blind tests, even though during "open" tests in which they were told correctly when a signal was absent or present these individuals exhibited physical and mental symptoms.
The researchers were criticized on a few grounds that they've now clarified; surprisingly, not all organizations that believe electromagnetic radiation can cause ill health thought the study was flawed. The primary critique was that 12 sensitive individual dropped out of the study before the double-blind portion. Critics suggested that those people were actually the true sensitives, and that the study was skewed against that point of view.
Researchers posted an extensive update on the matter in which they demonstrate that the characteristics of symptoms exhibited by the withdrawn subjects were very close to the 44 people who continued to the end. They note that five of the 12 withdrew because of their reaction to the exposure in the study, and two were excluded because they were taking medication that could have affected results. The first phase testing of these subjects was included in the final results, meaning that true sensitives, if they existed, didn't have markedly different symptoms from the subjects who completed the study.
Critics also pointed to the study's funding source, a joint effort by the mobile industry and UK government to research health effects. The researchers note in their FAQ that they their work was not subject to advance review after the grant was issued, nor did the funders have control over the release of results or their conclusions. The researchers note that their method of double-blind study removes results, too. That's true. With double-blind studies, you can critique the method by which bias was removed and information kept from participants and researchers during testing phases, and what's being tested--if the frequencies used or power levels involved don't match real-world conditions, for instance, that's a issue of the framing of the study and is a valid point of argument.
I contacted the research group, and on that last front, was told by Denise Wallace, a senior research officer, that the "power density used in our study 10 mW/m^2) was based on a recommendation from the National Radiological Protection Board" that's now part of the Health Protection Agency in Britain. These levels, therefore, are commensurate with what are considered safe levels found in the public sphere.
I asked Wallace whether the study results could be generalized to mobile phones themselves, as this study focused on the typically higher levels of power but more dispersed signal of base stations on "masts" or towers. She noted that their study was in line with other published double-blind studies that looked into matters such as human attention with GSM and continuous wave mobile phone signals present, and health effects from mobile phone usage on sensitive individuals.
Lufthansa is talking to partners about bringing back Internet service on its long-haul fleet: Lufthansa was an early and enthusiastic Connexion by Boeing customer, putting the satellite-based service that shut down last year in 65 of its 80 long-haul planes. That was an expensive proposition, costing hundreds of thousands of dollars per plane according to sources I consulted over the last few years. Lufthansa's customers also loved the service.
While Panasonic Avionics said in third quarter 2006 that they were looking to re-establish Connexion by using more efficient, lighter-weight technology that would employ the same satellites, they stopped speaking about it after their initial attempt to get 500 aircraft committed to the plan. They expected much higher bandwidth--12 Mbps/1 Mbps versus 5 Mbps/1 Mbps for Connexion--and thus a greater variety of airline-focused and entertainment services.
Lufthansa won't discuss its partners, and Panasonic wasn't mentioned in the Wall Street Journal article that broke this news today. T-Mobile's European division is looking to bring both Internet access and cell phone/data back to the air, and Lufthansa is apparently most closely aligned with them--although the Journal says the airline has no interest in the voice part.
AirCell announced its first customer this morning: American Airlines: AirCell won the coveted air-to-ground spectrum auction a year ago that gave them the right to a few scanty megahertz of spectrum over which they can deploy about 1.5 Mbps of throughput in both directions to aircraft in flight. Data service via Wi-Fi will be available; cell calling and cell data isn't in the mix given that the FCC decided to continue an 850 MHz cell frequency ban, and the FAA doesn't have a plan or a decision on the "airworthiness" of cell phones. We're safe for the moment, as AirCell can suppress VoIP service, too. (Update: Paul Robichaux blogged his conversation with American Airlines' in-flight communication and technology manager, which includes a lot of detail on the airlines' voice plans--or, rather, lack thereof.)
Unlike Connexion by Boeing, which had to invent the wheel and had massive annual satellite transponder fees, AirCell is expanding its existing network of ground stations currently used for general (private) aviation. They can also use much smaller antennas and a lot of off-the-shelf transmission gear (they're using the cell data standard EVDO Rev. A, for instance). The antennas are mounted on the underside of planes, reducing drag and thus the cost per flight to airlines. This dramatically reduces cost of entry, cost to airlines, and ongoing operating expense.
American Airlines has committed to testing, AirCell says, not yet actual deployment. Fair enough, because no such commercial offering has ever existed. The tests will start in 2008 and cover the entire area of the US; AirCell is negotiating for its service to operate in Canada, Mexico, and Caribbean, but had no announcements today about progress there. Previous radio-telephony services, like Verizon AirFone, were granted operational permission across those countries based on the US licenses.
The company told me this morning that passengers will have access to some content at no cost through a Web portal, including AA.com and sites that affiliate themselves with AirCell. Passengers would be able to book cars and hotels, for instance, both through American Airlines site, and through affiliated rental and accommodation sites. In the future, the onboard servers could also store movies and other media content, the company said, although there's no plan for testing that in this initial deal.
Pricing hasn't been announced, but AirCell reiterated previous statements that service will be at a slight premium over what hotspot service costs on the ground. I would wager they will look at $15 or less for flights of four hours or more; $10 for shorter flights; and $5 an hour for pay-as-you-go. They might also choose to offer a single flat-rate for all flights for simplicity, where pay-as-you-go would make sense on short flights, and unlimited on longer flights.
I would also expect that at the time of the test launch, they will have partnerships in place with aggregators like iPass and Boingo. That could be tricky because both firms recently switched to or revealed unlimited use pricing plans. With Connexion by Boeing, having a Boingo or iPass account gave you $1 off the service, and allowed you to use your same login. With AirCell's offering, it's unclear what the pricing relationship would look like.
TowerStream pioneered T-1 replacement with wireless broadband for businesses: That's not to say they were first, but they established the market, and have slowly grown to more and more major cities. MetroFi formally entering the business in Portland, Ore., for up to 10 Mbps wireless broadband thus isn't a giant event in the history of business-grade circuits. Rather, it's a significant move in metro-scale networks.
Service will start at $350 per month for T-1 speeds and rise to $1,500 per month for a 10 Mbps circuit and $1,000 for the receiver, the company told me. By comparison, Speakeasy Networks charges $390 per month for a wireline T-1. Towerstream charges $500 for a T-1 speed connection with bursts up to 3 Mbps, and $2,200 per month for 8 Mbps and $2,970 per month for 12 Mbps; their prices vary based on service level agreements.
As I and others have been writing over the last few months, there's been a kazoo-bassoon duet playing behind news coming from municipal markets. Subscribers numbers are poor. It's hard to get an indoor signal without a Wi-Fi bridge that costs $100 to $150 typically, but could run $300 for a higher-powered unit. And so on.
Thus the public announcement that MetroFi is actively seeking business customers looking to switch from wireline or gain far higher access speeds is good news--it's another line of revenue for the company. It's something that EarthLink et al. promised in the very beginning, too. Business-grade service using pre-WiMax, WiMax, or proprietary point-to-multipoint technology would be one of the revenue legs in the municipal network stool. All the metro-scale network buildings have to put some kind of aggregating point-to-multipoint hubs to carry their backhaul; the topology varies, but it's all there.
In MetroFi's case, their technology partner SkyPilot originally designed their networking architecture to handle lots of high-speed connections from pointed antennas (not omnidirectional ones). SkyPilot uses some clever techniques to create the equivalent of a dedicated wireless connection for very brief periods of time. (They have one radio and switch in a time-synchronized fashion among eight antennas, only broadcasting to a 45-degree arc and a single paired device at a time.)
MetroFi's release said that they could have service to a location in the covered area typically within a week. Upping the speed is a matter of a configuration change by MetroFi. MetroFi is also offering a service-level agreement; these contracts bind the provider to specific speeds and uptime, with discounts when those goals aren't met.
T-1 lines are the gold standard for reliability, although I have heard horror stories from plenty of telco customers that have seen nothing like 99.999 percent availability, and only excuses from the incumbent providers. Wireless broadband's advantage to a customer is that bandwidth can scale above 1.5 Mbps with no additional equipment or installation. The leap from T-1 to fractional T-3 (about 45 Mbps, sold in increments up to that level) is rather large. Broadband wireless can also be installed in as little as minutes, in a pinch, and the provider has no wirebase to depreciate, just equipment on either end; they typically get the customer to pay for their receiver, or get a one-year or more commitment to eat the cost of the receiver.
Thirteen airports are covered in a free Wi-Fi for iPhones promotion: Boingo Wireless is reaching out to the hundreds of thousands of we early adopters with iPhones by offering free access in 13 airports. They purchased Concourse Communications last year, the leading airport Wi-Fi vendor. Boingo charges $8 for a day session or $22 for unlimited US and $39 for unlimited worldwide access. No special coupon codes are needed: Just select the "concourse" network from the list of Wi-Fi networks and connect.
Because the iPhone doesn't yet support third-party applications, using Boingo requires the entry of your user name and password on a login page each time you want to connect. This promotion should let iPhone owners test Boingo's network with less friction; ultimately, Apple will allow outside software, and a Boingo connection manager would be the first thing I'd want on mine.
The airports are Baltimore/Washington, Detroit, JFK, Kalamazoo/Battle Creek, LaGuardia, Lambert-St. Louis, Midway (Chicago), Minneapolis-St. Paul, Nashville, Newark, O'Hare, Toronto, and Will Rogers (Oklahoma City).
(An enterprising soul might recognize that the iPhone is recognized by an HTTP referrer string sent by its Safari browser.)