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 Joint Venture Silicon Valley public/private partnership has issued its RFP: The group of cities, counties, governmental bodies, and corporations want a wireless network of some kind--technology isn't decided and could be a broad mix--that would cover Silicon Valley. Winning vendor(s) will be selected from the respondents to their RFP by September, and recommended to the 16 cities, San Mateo County, and 16 other jurisdictions that have signed on.
I wrote in January about the scope and nature of this 1,500-square-mile potential project.
Worcester's mayor wants Wi-Fi, expanded commuter rail: Mayor Timothy P. Murray is running for lt. gov., and he wants more statewide commuter rail with onboard Internet access an important component for encouraging use. He proposes the state buy freight rail tracks to gain more control over use.
Austin's Wi-Fi network's first phase up and running: The network was built for an upcoming information technology conference that will leave behind the nucleus of what the city wants to have become a metro-scale network. Cisco donated the early equipment, which comprises 28 access points.
Orlando plans 9 sq. mi. network: Despite being cited as a failure by another city in Florida a few days ago. The service will be free, throttled to a certain level to be determined, but which apparently could be as low as "two times dial-up" (128 Kbps?). The firm Smart City will pay something like a franchise fee--6 percent of gross revenue--to operate the service.
Tempe's University of Arizona will build its own network: Despite earlier speculation that NeoReach would extend its Tempe network into the UA campus, this story makes it sound as if the university will build its own network using student fees. The cost will be $65 per year per student and $6.9m overall. My mistake: The dateline is Tempe but the University of Arizona (UA) is in Tucson; Arizona State (ASU) is in Tempe.
Montreal suggests waiting to see what other cities do before leaping into metro-scale service: Montreal's community wireless group has installed 70 hotspots, and the city is apparently in no rush to be the schlemazel who suffers the missteps of a pioneer. The city has no timeline for such a service, and one of the founders of Île Sans Fil (that would be "island without strings" or "wireless island") thinks the city should follow their model. Update: See comment below. The fellow quoted in the story from the community wireless group says the reporter mischaracterized his opinions.
Boca Raton's city council puts Wi-Fi at bottom of priorities: The council would prefer a paid system to bring in revenue, and points to what they characterized as an expensive and failed downtown Wi-Fi test in Orlando. The assistant city manager cites a few cities he says that have installed "hot zones," but it's an odd selection, and most of these kinds of hot zones are paid for with private money (downtown chambers of commerce or privately funded by ISPs). Other statements appear a little odd, too, frankly.
Pittsburgh pauses: While it seemed ready to sign a utility pole agreement with US Wireless to provide a combination of paid, limited free, and digital divide Wi-Fi service citywide, the city council had an odd vote (three for, three abstain, three absent) on a proposal they needed to sign off on for the deal to proceed. They may vote again on Tuesday. One abstainer said he wants a term shorter than seven years. There's also reportedly concern that US Wireless's deal has incentives that make an unlevel playing field. EarthLink has apparently offered to build a network with no incentives. A few days ago, I wrote about "Muni-Stall," or the apparent delays that emerge when municipal decisionmaking is engaged. Elected officials often make decisions slowly, regardless of early quick progress, and that's not a bad thing when long-term contracts (five to 15 years, often) are involved.
Suffolk County considers unwiring 900 square miles: There are 1.5m residents of this New York county which would bid out for a private firm to bear all risk and cost associated. It might include pushing service out over the water, too, given the extensive coastline. Suffolks is an odd mix of non-urban landscape, this article in The New York Times explains, which may make it appealing for Wi-Fi coverage. A local Wi-Fi/networking guru, Craig Plunkett, wrote a response to an early floating of this plan almost three months ago.
Let's trust their reporting, but it's an informal record: A team in Venezuela went up a mountain and came down the unofficial holders of the longest Wi-Fi link record at 167 miles. They used Linksys WRT54G using open-source firmware (DD-WRT) using retrofitted satellite parabolic dishes.
When I read this story initially I thought, well they may have used Wi-Fi protocols, but this isn't Wi-Fi in the sense of regulation-conforming unlicensed use within the signal limits. The last Wi-Fi distance winner, Cincinnati teenagers who created a 125-mile link for DefCon's Wi-Fi Shootout, clearly exceed Part 15 rules, but claim they were using Part 97 (amateur radio) rules. Part 97 ain't regular Wi-Fi.
But I ran the math for the South American claim of 279 kilometers (167 miles), and it seems to check out under FCC Part 15 rules. I don't know whether Venezuela is regulated in the same fashion, however.
They said they were using 100 mW output from the Linksys WRT54G with a very short cable to reduce signal loss. Their antennas they calculate at 32 dBi, although that's a rough number. That gives them an extremely high EIRP (effective isotropic radiated power) of 158 W, which seems very high. I went to look up the FCC Part 15 rules and found that rule 15.247 allows a 30 dBi antenna with a 22 dBm or 158 mW power-at-antenna signal. With no amplifier between the Linksys and the antenna, I believe they qualify there. So they might be slightly over on the antenna gain, but this allows them an EIRP of up to 158. So far, so good.
The next question would be link budget: with two identical setups over the 270 kilometers of the test they set up with an unimpeded Fresnel zone--no obstacles through the entire spread beam between the two locations--they have a pretty clear formula. Assuming 20 dBm (100 mW) at the Linksys, a very small -3 dB cable loss, two 30 dBi antennas, and a 149 dB of free-space loss, with a receiver sensitivity of -85 dBm (which checks out with what they saw)...they have a 10 dB link budget. Which should be sufficient for a test.
If any of these numbers are off by a relatively small percentage, they'd be violating the law or the test wouldn't work. But it's a fairly impressive link at the edges of what (at least in the US) is legal and certainly what's possible.
I'm not an engineer, and thus I may have made errors in my assumptions. Please add notes in the comments if you'd like to correct my math or figures. (I used this WLAN Link Planner for link budget, this reference for Part 15.247, and this EIRP calculator.)
The EasyShare V610 uses Bluetooth 2.0 for wireless file transfer: The 3 Mbps Bluetooth radio is built in, and accompanying software allows photos to be transferred to a host computer and from there uploaded to Kodak's EasyShare Gallery (formerly Ofoto). The $449 camera has two lenses, like the earlier V570, to provide a remarkable range of optical zoom in a very small form factor (38 to 380mm with a gap between lenses in the short-range zoom from 114 to 130mm). It's a six-megapixel camera, also quite stunning for its 4.5-by-2.25-inch size. It ships in May.
Buried at the bottom of a press release--which is not on Kodak's site yet--is a note that the next model of the EasyShare One, Kodak's Wi-Fi capable camera, will ship this summer with a set of profiles for wireless ISPs, as they call it, which means hotspot networks. They originally shipped the camera with a T-Mobile connection option. The new model will cost $299 with a $99 Wi-Fi card as an option; this is down from the $599 cost of the original, which included Wi-Fi. Confusingly, the first model was supposed to cost $499 with Wi-Fi as an extra, and then Kodak decided it wouldn't release the camera without Wi-Fi. (I reviewed the camera last October.)
I found the EasyShare One to be a great but overpriced camera with irritating Wi-Fi connectivity options because of the many steps, frequent slow reconnections, and lack of WPA security. Kodak slipped out a firmware upgrade at some point (no date is on the upgrade) that offers WPA Personal and other options. This upgrade was promised for some months ago, so I don't know why they were quiet about releasing it.
This week's St. Cloud, Fla., story about some minor glitches received big play: While the Associated Press story was balanced and had a lot of information about the limitations of that free city-wide network, most of the analysis of the AP story relied on the first few grafs which mentioned the experience of one individual who couldn't get a reliable connection. This isn't unusual, although it's a problem with the echo chamber that is second-hand Internet analysis that the gist of the story is lost through focus on one facet of the reporting.
The St. Cloud story isn't important in the grander scheme, however, because St. Cloud managed to build their network and take it operational. While Esme Vos's latest count of metro-scale networks--municipal use only, public only, and mixed--shows 193 cities in the US with operational service or issued RFPs, the scale remains small.
A colleague recently complained via email that everyone keeps talking about Chaska, Minn., the small town that used Tropos equipment nearly two years ago to provide ubiquitous broadband coverage. Chaska is an interesting story, and appears to have been generally successful in its cost and service goals. But it's hardly a microcosm of the larger networks that are being planned and built in urban areas.
The stall has to do with the pace of government, something I've wondered about for more than a year. With the speed that some RFPs were issued, it seemed like the usual pace of deliberative government was being bypassed. This didn't seem like a good thing for two reasons.
First, with public funds (let's eschew taxpayer dollars, a loaded and incomplete phrase) involved, there should be plenty of time for review and for critique of business plans, some of which seemed too enthusiastic to start with--assuming that incumbents wouldn't lower prices or enter the market. Even when public funds aren't used to build the network, almost exclusively the case in the largest markets in the last year, public dollars will be used to pay for telecom services contracted with a winning bidder. And the winning bidder will receive a certain kind of franchise and exclusive right as the first mover and the bid winner. These quasi-franchise contracts have 5, 10, and 15 year terms.
Second, technology is changing so rapidly that building a network in 2006 may not make as much sense as having built it in 2004 or 2005. It's hard to make a case today that a network you start work on this month will have the best speed, performance, cost advantage, and spectrum efficiency as one that breaks ground in a year. With 802.16-2005 (the ratified version of 802.16e) and mobile WiMax entering the picture for niche purposes, including a specific citywide service overlay, and with 802.11n swinging towards ratification, it seems like a hard argument to say that vast speed and coverage improvements won't be available for metro-scale devices in the next 12 months. (The recent public launch of Go Networks, which uses MIMO to reduce metro-scale node count by up to 50 percent, the company states, is one indication of the direction that at least claims in the market will trend towards.)
The latter factor probably won't prevent networks from being built, because companies are bearing the financial risk themselves of having to later swap out equipment, and are meeting specific speed requirements in RFPs often composed to avoid direct competition with wireline DSL and cable providers. Most of the RFPs and the most of the bidders and providers involved in those metro-scale networks already include components in their networks that should allow some futureproofing, such as WiMax-like components for backhaul (which will be used for T-1 replacement for muni and business users).
But the former factor, the deliberative pace of government, has become a major issue. I notice that Philadelphia is often cited by journalists around the US and internationally as a network that's in the process of being built. Or a reporter will write that Philadelphia's network is a model for other cities. This is, of course, a canard.
When we look at large cities and the networks that are being built for them, you can see a pattern.
As I note above, there's nothing at all wrong with this taking some time. In the case of SF, one person notes that the contract that will be signed lasts 15 years, but doesn't require an increase of bandwidth during that period. “ 'In 2021, 300Mbps [sic: should read Kbps] is going to seem a bit ridiculous … it's a great solution for, like, 1996,' said Ralf Muehlen, director of SFLAN, a free, nonprofit Wi-Fi network in the city."
When I wrote the article for The Economist that was titled Wi-Pie in the Sky, I concluded that no network of the scale, density, and nature of the one proposed for Philadelphia had yet been built. Mesh vendors gave me pushback on this statement while I was writing the article and after it appeared. But none has been able to find an example that has more than one or two of the several properties that all these major city networks will have to demonstrate.
What this means is that most of these networks will be built simultaneously and not be able to learn from mistakes or successes of the others during testing and build-out. I had thought months ago that the 15-square-mile test network that Phila. required the winning bidder to create would have been in operation for some time. Now it seems that the pace of government wins over the pace of technology in the end.
That could be a good thing. Enough delays could give time for plans to be revised to incorporate 21st century technology instead of good old last century's 802.11a, b, and g.
Annapolis Wireless will offer service starting downtown: The Nortel-powered network will be free with corporate and non-profit sponsors already signed up to pay expenses. Use requires registration. The city sees 5.6 million tourists a year.
The press release contains an unfortunate paragraph:
"The creation of a city-wide WiFi system marks Annapolis' entrée into a technologically sophisticated group of cities. Cities such as Philadelphia, Houston and New Orleans also provide free WiFi access. But Annapolis' WiFi carries a distinct difference - many other cities provide the access using taxpayer funds. Annapolis Wireless, by offering sponsorships and advertising to area businesses on its splash page, is able to provide WiFi access to the public without using taxpayer dollars."
The vast majority of citywide networks involve no public funds, and the folks behind the Annapolis network didn't need to make an over-the-shoulder nose tweak at other cities in this way, it seems. The big difference between Annapolis and many other cities that offer or plan to offer Wi-Fi is that most companies providing that scale of service think free isn't sustainable or at least requires advertising (not just sponsorship).
The Belen to Santa Fe commuter rail route will be equipped with Wi-Fi transmitters: It will likely be free. Albuquerque, between Belen and Santa Fe, has free Wi-Fi at its airport and on 12 buses. The rail Wi-Fi will hook directly into fiber-optic lines that parallel the tracks. They're estimating 200 access points to cover the approximately 100 miles of track. Limited rail services begins in June with the full line in operation by 2008. The rail authority will bid out the Wi-Fi service operations.
A private firm in Melburne wants to install Wi-Fi on Victoria's public transportation: They'll equip bus and train routes with free service while displaying information and advertising. The excitingly named 802! has been testing service, and can provide seamless handoffs at 100 km/h. There's a 500 KB file download limit from 4 am to midnight.
Three buses in Cincinnati will have Wi-Fi hotspots onboard: It's a one-week trial, but could become permanent.
[Please note: This editorial was written in 2006. In 2007, I changed my advice when Draft N was finalized into version 2.0, with changes incorporated into shipping hardware. Then the Wi-Fi Alliance added a pre-ratification certification process. I fully recommend Draft N gear now. –Glenn Fleishman, 2009.]
I am going to make myself extremely unpopular this morning by suggesting that no one buy so-called Draft N Wi-Fi gear that is pouring into the market: Buffalo and D-Link say they've been shipping wireless equipment based on the current draft of 802.11n, the higher-speed cousin to existing Wi-Fi, for at least two weeks. Linksys entered the fray today with a new router and PC Card. Other manufacturers who have not yet announced will follow in May and June. And it's a bad idea.
These companies are shipping Draft N devices for the bragging rights to be first out of the gate and to try to brand their Draft N products' identities on consumers' minds. There's no good technical reason to release these products this early. Lest someone point to 802.11g's early release before ratification by the IEEE, it's important to note that 802.11g was several drafts further along when Broadcom pushed out its first chips.
I came around to the idea of early multiple-in, multiple-out (MIMO) equipment more than a year ago because MIMO gear, once it started to drop in price, offered the significant advantage of dramatically increased coverage and range even while using older Wi-Fi adapters. For specific purposes in which you needed more throughput for a wireless link, purchasing all new MIMO gear also made sense as a limited investment with no path to ensure that the higher speed of that equipment would work with any future standard. So far, so good.
But Draft N gear makes no sense for consumers because the actual standard is close enough at hand that buying pre-standard equipment has little benefit and a possibility for disruption given manufacturers' current guarantees and claims. Here's the lay of the land.
No guarantee for upgrade. There's still a small but significant risk that equipment shipped today will contain chips and firmware that cannot be upgraded to provide full compatibility with the final standard. Buffalo was brave enough to say in an interview two weeks ago that they are not guaranteeing that their Draft N equipment will be upgradable, although they believe it almost certainly will. Other manufacturers are not so brave, but they are also not offering a replacement guarantee.
With no hardware replacement guarantee from a manufacturer, why buy today? If manufacturers are willing to step forward and provide explicit offers with their products that they will have the same compliance with the gear they shift today as with the gear they ship in six months, up to and including replacing the hardware you purchase today, then that argument goes away.
No agreed-on, tested plan to avoid interfering with older networks. A more significant concern is that chipmakers and the IEEE task group working on 802.11n haven't agreed on a single approach to mitigate interference with existing legacy Wi-Fi networks, nor are the individual methods well tested ina real-world environment or with each other.
Over a month ago, I wrote about Airgo's self-motivated concerns in pressing this point, but I'm now convinced that they're generally correct about the issue, although they certainly overstated some of the particulars. Without agreed-upon methods of avoiding interference, many different chipmakers and equipment manufacturers devices will be contending with an inevitable variation in success.
The idea that consumers will apply several firmware upgrades to solve problems that emerge from this contention is a little ridiculous. Although early adopters will likely buy this Draft N gear, there's a lot of evidence to suggest that a device, once deployed, is rarely patched unless someone is advised to do so by technical support or they're a fairly technical user to begin with.
Testing of the first Buffalo and NetGear devices by Craig Mathias of the Farpoint Group, showed poor interoperability, security model problems, and performance. Mathias also recommends avoiding this wave of draft products. (News.com picked up this story and pointed to eWeek's evaluation showing poor range in a test device and legacy network problems, too, but praising the potential. They also noted that Linksys will release the same product with Atheros, Broadcom, and Marvell chipsets depending on availability. Hello, tech support nightmare.)
Gigabit Ethernet is necessary for the full benefit. If you're planning to buy 802.11n equipment for coverage area, this isn't an issue, but if you're not buying it just for range, you need a gateway that has Gigabit Ethernet (1000 Mbps) built in. Only NetGear is offering models that support this speed in these early stages. GigE has become less expensive, but it will take market interest and some months before it's available for a reasonable price.
Prices will plummet within months. If you buy Draft N gear today, you're paying too much. Five major Wi-Fi chipmakers will contend for business in Draft N. Airgo said they're waiting a bit longer to release chips and Intel's plans are unknown, but Atheros, Broadcom, and Marvell are already producing silicon. With five chipmakers and dozens of manufacturers, competition will be intense and prices will fall fast.
Wait. Unless you have a very particular, very short-term need for the highest possible speed using a wireless network, there is little reason to suffer through the shakedown that will take place over the next several months. At the very least, wait until after the May IEEE meeting, at which point it may become crystal clear whether any minor changes to the draft of 802.11n will require major changes in silicon. Even if there are no chip changes, you might as well wait for the inevitable firmware upgrades that will chase the standards group meetings and take weeks to appear following those meetings.
I won't recommend buying Draft N gear until almost certainly September or later--later if the prices haven't fallen substantially by September.
Education may be the key to launching new Wi-Fi networks: St. Cloud, Florida, is suffering from what appears to be a backlash against its citywide free Wi-Fi network given that it's only been in operation a month. The service obviously--to those of us that follow this industry--can't reach every resident or point in town, especially not in the first roll-out, and yet those who are uncovered in this 28,000-resident suburb of Orlando are apparently hopping mad out of the gate.
As I'm quoted saying in this Associated Press article, there's an important element of education that new networks need to spread to potential customers, even when those customers aren't paying anything to use the network. We all know that the most aggravated people make the most noise, and that with Wi-Fi, it's very easy to get a marginal signal that's frustrating because you can see a network and yet not reliably connect to it.
For instance, one resident quoted throughout the article says, "I don't think it's going to work. Very few people are going to use it, and they're going to say it's underutilized and they're going to shut it down." But he's reasoning from first-hand experience. HP, which built and is operating the network, reported 50,000 users sessions in the first 45 days and just 842 help line calls. The service has seen 3,500 registered users and 176,189 hours of usage.
But it still sucks to be the person in an area of awkward coverage. The city is offering a bridge of some kind for $170 to improve weak signal areas, but that's quite high. A $50 to $80 bridge should provide the same benefits.
Peter Judge writes a UWB Soap three-parter: It's daytime theater on the air as Judge presents the views of Freescale, the Bluetooth SIG, and the WiMedia Alliance on the future of UWB and Bluetooth. You'll laugh, you'll cry, you'll consult stock market reports.
Part one covers the views of Martin Rofheart, a pioneer of UWB and head of Freescale's wireless efforts, but whose particular flavor appears marginalized and sidelined to my eyes. Freescale's UWB technology was dropped by Bluetooth and Freescale and former parent Motorola left the trade group they helped form. While promising shipping silicon for years, Freescale now seems poised to deliver it at last this July--but only running at 110 Mbps, not the promised 480 Mbps speed sought for the last couple of years. Rofheart makes a lot of claims for how his competition's version of UWB is further behind and less capable. Freescale plans to deliver Cable Free USB, which will work seamlessly through dongles and hubs with existing USB 2.0 connections and devices.
Part two looks into whether Bluetooth has a future given recent developments. Judge notes that Freescale's Rofheart believes Bluetooth over UWB will be late to market and have a restricted worldwide appeal due to choices about spectrum. Will no-wire USB simply take the day? Hard to say, although there's a lot of good technical detail here to absorb. Fundamentally, Bluetooth is about applications running over a radio, while USB is about connectivity. For companies already invested in Bluetooth, adding UWB as a radio option should involve less effort than retooling around USB. Although USB is commonly used for phone synchronization, too. It should prove entertaining.
Part three has the WiMedia Alliance chiming in with their upcoming delivery dates and refuting much of what Rofheart says. (There's no rebuttal from Freescale yet, but part three just posted.) The WiMedia Alliance says that Certified Wireless USB will provide better throughput than Freescale's Cable Free USB by a long shot, and that a single radio for future application and connectivity standards for short-range networks is the right way to go.
The soap opera won't reach its conclusion until products ship. Stay tuned.
Yuma signs with NeoReach: Another Arizona city falls to the persuasive power of this southwest powerhouse division of MobilePro. This is the fourth city in Arizona that NeoReach will unwire. They'll cover 25 square miles of the 107 sq. mi. town, and exchange free access for city employees in lieu of pole and electrical fees. Yuma sits on the border with California.
A stupid, stupid law has been passed in Westchester County: Sometimes righteous technical indignation overcomes me and I must, like the mad prophet of the airwaves in Network, tell you to throw open your DVD drive doors and launch Windows and shout, "I'm firewalled as hell, and I'm not going to encrypt it any more!"
Months ago, word emerged from the tony suburb of New York that is pony-infested, mansion-encrusted, millionaire-swarming Westchester County that a law would be passed requiring businesses collecting personal information that use Wi-Fi to have minimal protections, and businesses that offer Wi-Fi to post notices about security.
Seemed a little silly to try to regulate this at a local level, but when you read what they actually planned to do, it was apparent that no one involved with the law understands precisely what they're asking for: they're trying to regulate a solution that only covers part of the problem. By misstating the security risks, they're not serving their constituents.
Months have passed and so has the law, which goes into effect in 180 days.
The law correctly describes a firewall: "Firewall" shall mean a set of related programs or hardware, located at a network gateway server that protects the resources of a private network from users of other networks.
A firewall only protects computers from outside threats, and then only if placed at the correct point in the network's topology (at all entry points for outsiders). It does not protect against the interception of data passing across the network. A firewall is a necessary first line of defense for any company network, and many Wi-Fi gateways include adequate to great firewalls that employ one or more well-acccepted techniques, starting with network address translation (NAT), which creates private, non-routable addresses, all the way up to active firewalls with packet inspection that recognize and block well-known attacks.
The law as enacted could help educate businesses to this particular threat: the invasion of their computers via Wi-Fi networks that they operate. Properly configured and placed--and viruses, worms, and other malware separately invading the network aside--a firewall could prevent private data from being obtained via a Wi-Fi network that otherwise would allow direct access into a company's wired network.
Westchester appears to believe the firewall solves the whole security risk caused by using Wi-Fi. It doesn't solve what's typically seen as a more significant problem. Wi-Fi data, when unencrypted either by a Wi-Fi security protocol or by a separate encryption session (like a secure Web session), is passed in the clear to any other user of the same network.
The chief information office of Westchester County is paraphrased in this article: "Jacknis said easily available firewalls would protect credit card transactions, for example, from being detected by a hacker posted outside a dry cleaner that uses a wireless network."
The firewall might protect a hacker from gaining access to computers running credit-card transactions. But if the computers at the dry cleaner's were connected to the Internet or to each other via Wi-Fi and they don't have encryption of some form enabled, and the credit card transactions aren't encrypted (which they should be, of course), then those transactions are freely available to any hacker.
The tips for securing Wi-Fi networks are weak, starting with changing the default SSID or network name and disabling SSID broadcast. Down the list of suggestions on improving security, there's a small mention of enabling encryption.
Companies running public hotspots have to firewall their own machines against the open network, as I read the law, and have to post a fairly dopey message that's not an accurate statement of what's at risk.:
YOU ARE ACCESSING A NETWORK WHICH HAS BEEN SECURED WITH FIREWALL PROTECTION. SINCE SUCH PROTECTION DOES NOT GUARANTEE THE SECURITY OF YOUR PERSONAL INFORMATION, USE YOUR OWN DISCRETION
The sign should more accurately state:
ALL THE DATA YOU USE ON THIS NETWORK CAN BE RECEIVED BY ANYONE ELSE ON THE SAME NETWORK. USE A PERSONAL FIREWALL AND USE SECURED CONNECTIONS FOR WEB BROWSING, EMAIL, AND OTHER SURFING. OR RISK THEFT OF PASSWORDS AND PRIVATE INFORMATION.
That's a little more direct, no? It's accurate but so frightening it might drive off all hotspot users.
More on the 802.11s mesh standard as word of a January compromise hits trade publications: Apparently, we all missed the news in January and March that progress was made on 802.11s, a mesh standard coming through the IEEE process, because Motorola's announcement a few days ago that they would support the draft and final version of the standard has sparked several articles. At Wi-Fi Planet, Eric Griffith tries to figure out the scope of what the spec encompasses. This can be tricky, because if you're outside the IEEE (or inside and not yet a voting member), you don't have access to the draft itself.
There seems to be a bit of back-and-forth over whether multi-radio outdoor nodes will be accommodated by the final 802.11s. Phil Belanger, who wrote about the indoor, peer-to-peer implications of 802.11s in a white paper published at Wi-Fi Networking News last month, thinks that the impact will be limited to end nodes that can act as mesh devices among each other.
Representatives of Motorola, Strix, and Belanger's former employer BelAir stated in Griffith's article that the implications are broader and that the standard now encompasses outdoor, multi-radio units. But none of the firms would suggest that the standard would, in fact, allow fully interoperable outdoor multi-radio nodes! In the best case, they'd have some minimal level of interoperability that would probably not be terribly useful.
Thus Belanger's original contention seems to be the accurate one, then: in practical terms, it's niche and end-user nodes that will benefit from 802.11s.
A story from Oakland County, Michigan, shows the power of utilities in the future of metro-scale Wi-Fi, WiMax, and what have you: MichTel is to blanket over 900 square miles of Oakland County during a two-year build out that will involve what's reported here to be as many as 20,000 access points, but which the company expects to net out to be fewer as technology improves during their deployment.
The stumbling block they appear to not have reckoned with is the utility, DTE Energy, which owns most of the utility poles on which the access points will be mounted. Oakland County, this story says, is providing access to just 2,700 mounting points outside of the utility's poles.
Oddly, MichTel comes off as quite naive about a utility's interest in providing speedy access to their poles. MichTel's general manager said in this story, "DTE said we have to identify the pole, send the information to them. They have to send someone out to look at the pole. They haven't been able to give an answer on how long that will take. A day? A week? Weeks? All that impacts being able to start and finish in a reasonable amount of time."
Part of the Telecommunications Act of 1996 as interpreted by the FCC specifically requires that, in the absence of good-faith negotiations that proceed first, "a utility charges just, reasonable and nondiscriminatory rates for pole attachments." It's a pretty involved interpretation (read up on the overlashing provisions, for instance), and it primarily sets the groundwork for a way that an entity that is unable to get access to poles can complain to the FCC and have orders and sanctions issued against the offending entity.
Because DTE Energy and other utilities don't routinely make available thousands of poles at a time, it doesn't seem unusual at all that they have neither a timetable nor an expedited process. In some cities, the requests for proposals (RFPs) for wireless networks lay out the attachment issues, describing what poles, buildings, and other facilities are available for mounting. Some cities own the utilities, which may make it easier for blanket contracts to be signed. In Philadelphia, three of the five contracts that EarthLink and the city have to sign involve utilities and mounting fees.
And, note above, the 1996 Telecom Act's order on poles only requires it be used as a backstop to negotiations that fail. A Wi-Fi or WiMax operator could negotiate a deal with a utility better than that defined in the FCC order. Pole access fees aren't meant to provide real profit to the entity that owns the pole, but to make it worthwhile to comply with the rules.
When Tacoma Power began work several years ago to build their fiber/coax infrastructure that became the Click! Network, they voluntarily ate the cost of replacing thousands of poles, and moved incumbent telco and cable wire at the power company's expense. Why? Because it was worth it to them to have the best pole infrastructure to support their new system, and to replace inadequate and patchily maintained poles. But this demonstrates the power of the incumbent, too, to make that sort of scale of activity happen when it's beneficial to them as well as their pole tenants.
Access to facilities and rights of way are what defined the franchise rules for cable companies to provide television service across the way, and franchises are perpetuated because they bring in recurring income to towns and cities. These new metro-scale wireless networks are a new form of de facto franchise in which the new operators will pay recurring fees and taxes of all kinds. Whether utilities will impede or streamline networks that rely on their infrastructure is an open question.
Airport Wi-Fi operator ICOA released its latest securities filing, and it's interesting reading: The pink-sheet-traded firm's 10-KSB form shows that their revenue is way up--but they have enormous accrued debt and liability, too. The largest hotspot operators in North America are either private (cf., Wayport, Surf and Sip, Fatport) or part of larger entities that don't break out hotspot revenue, expense, and funding (cf., SBC-aka-AT&T, T-Mobile). ICOA is thus the largest firm which provides primarily hotspot access for which I have seen actual financial filings. (Please note that this post doesn't constitute financial or investing advice of any kind.)
The company's fiscal 2005 revenue was up 111 percent over FY2004, from $1.2m to $2.5m, which is a significant uptick, representing the continued maturity of the airport Wi-Fi market and the partnerships they have with aggregators for resale. However, the firm reported a net loss of $9.2m, although that includes $5.3m in one-time costs and accounting charges.
Their statement shows a total of $19m in capital inflow but due to -$23.4m in accumulated deficits, the stockholders' deficit is -$4.2m. Their current liabilities also far outweigh current assets. A statement on the viability of the firm as a going concern notes, "The Company needs to raise a minimum of $2,200,000 through public or private debt or sale of equity to continue expanding communications services, voice, facsimile, data and electronic publishing network and the service operation center, and to develop and implement additional contracts at airports, hotels and retail locations in order to continue placing terminals in high traffic locations."
While these statements and numbers are interesting, I'd have to say that revenue and growth are the most fascinating part of this story. ICOA has grown through partnerships with 2nd and 3rd tier airports around North America that have significant enough traffic and yet small enough operations that it's likely both the capital investment and operational costs are much lower and the margins higher than at the 1st tier airports. They've also purchased a number of firms in the same or related businesses, expanding their ability to leverage fixed costs across a larger footprint and array of services.
While the deficits and liabilities are large relative to current earnings, their upside on earnings with current locations is much, much higher than they're seeing now based on passenger numbers and other factors. This note in the report shows that level of growth: "As of January 27, 2006, our airport footprint has expanded by 258% compared to the same time last year, from 7 to 25 facilities with a related 205% increase in annual passenger coverage from 20 million to over 61 million."
There's no way to predict the success of a firm that is leveraged so highly on execution--keeping the Wi-Fi going, keeping partnerships working, keeping technical infrastructure intact, handling billing and complaints. But these numbers at least provide some guidance for those wondering how much it costs to build Wi-Fi networks, anyway.
In the Great North, the small capital of Iqaluit has free Wi-Fi: The relatively new territory of Nunavut comprises 2 million square kilometers of land and sea and under 30,000 people; it was established through a long process that resolved generations of native peoples' land claims. The capital has about 7,000 residents, and a local telecom firm has put in free Wi-Fi using a well-placed antenna on a company building.
ComGuard CTS launched the system April 6, and will throttle connections to a low-enough rate (just 96K with a 2 Mbps connection overall) that they don't compete with wired or satellite Internet service that's available in town. They're working with residents and businesses to install repeaters and bridges to improve coverage.
While the Globe and Mail story says the firm is a two-man shop, one of the owners posted a note in the comments that they're larger--it's just two staffers were tasked with making free Wi-Fi happen.
For some fantastic photos of the setup and the landscape of this distant town, see the ComGuard discussion thread.
Business 2.0's Owen Thomas does the math on EarthLink's earnings: The company is still profitable, but is facing small increases in a marginally profitable broadband market (due to incumbent ownership of DSL and cable lines) and large decreases in its dial-up customer base. The company is working hard on quickly building a Wi-Fi infrastructure business, signing (or trying to sign) contacts with cities all over the U.S. to produce a new structure in which they own infrastructure and wholesale it out. Note to incumbents: EarthLink shares.
Analysts question EarthLink's ability to run facilities-based operations, although from my talks with EarthLink, it's clear that they're contracting out the physical part (which is out of their ken) to experienced telecom infrastructure partners, like Motorola. The logical facilities operations is in-house, and that's something they've run since their inception--billing, network operations, services, etc.
The conclusion? "As the cash that flows from dial-up shrinks, for EarthLink, building its own broadband network isn't an experiment - it's a matter of life or death."
MetroFi was given a contract by Foster City today: The four-squale-mile town of 30,000 will pay not a cent for MetroFi to provide its free (ad supported) or fee (no ads) service.
My colleague Cyrus Farivar put together this radio item on municipal Wi-Fi worldwide: The program that commissioned it wound up not having room to use it, but they provided him a produced and mixed version that's worth listening to. His blurb:
"Over the last few years, wireless Internet access, known as WiFi, has become more and more widespread. Various cities, ranging from Philadelphia to Paris are currently building citywide WiFi networks. There are other efforts to wirelessly connect whole countries as well, like in Macedonia and Estonia. But each country’s path is different."
Farivar interviews relentless Wi-Fi promoter Veljo Haamer, who is more responsible for the spread and availability of Wi-Fi in Estonia than anyone for their own country elsewhere in the world.
Motorola's Mesh Networking Group (Motomesh) has committed to supporting 802.11s: They say their current solution is draft compliant, although the IEEE task group S's notes from the March 2006 meeting states that a merger of two outstanding proposals was successfully made and "confirmed unanimously at the March meeting as the starting point for the 802.11s standard, although much work remains."
Motomesh is confident that their architecture allows over-the-air updates to whatever the draft resolves itself to be. It's unclear why being the only company to be draft compliant before "much work" has been processed on the draft is a selling point over simply stating that the products will be upgradable to the ratified version or later, more fully settled drafts.
As Phil Belanger, formerly of Wayport, Vivato, and BelAir, noted in a white paper published on this site a few weeks ago, 802.11s may have a lot of impact in devices meshing across homes or outdoor environments, rather than any impact on interoperability of mesh nodes that form metro-scale or enterprise-scale networks. It's about the end devices rather than the core mesh nodes, in other words.
Motomesh seems to agree.
TechCrunch shows a survey screen that asks the subject what they'd think about Yahoo! Messenger On-the-Road: The service would potentially be offered at over 30,000 hotspot worldwide for $7.95 per month flat rate or $2.95 for two hours of Yahoo! Messenger usage. The 30K hotspots and $7.95/month rate matches Boingo's statistics and its offer for VoIP with Skype, so perhaps Boingo would be the platform behind this.
It's an odd trend to see specific services being offered, such as the Nintendo DS Wi-Fi connection (which Nintendo offers at no cost, subsidizing it for its buyers), rather than generic Wi-Fi for all applications. It's about the stupid network, isn't it? But this isn't so much about smartening up a network as striating it for the purposes of providing a better rate for a specific activity, like VoIP.
Ricochet came up most recently a few weeks ago when an article appeared about its current owner's latest attempts to extend service: While Terabeam has 20,000 users in Denver and 4,000 in San Diego, according to the company, they were hamstrung by the way in which Metricom stopped operating when it went into bankruptcy. They left a lot of peeved municipalities in their wake. News a few days ago indicates the Terabeam is considering selling Ricochet, finding a partner, or obtaining investment. [link via TechDirt]
Ball State University will unveil a digital sculpture tomorrow that derives patterns from the campus Wi-Fi network: Giants screens will have projected on them "sound, color, patterns and images"--let's hope not the usual images that college students are viewing. It's a neat idea, though, to represent the nature and quantity of traffic as something you can see. You can view the stream of material online, too.
The airport has an interesting history with Wi-Fi that I've been writing occasionally about since 2003: It's a fairly small airport, not atypical for state and province capitals that tend to be located in politically expedient places that aren't often also bustling metropolises compared to the big towns that developed in their political unit. (Olympia? Albany? Austin?)
Sacramento originally contracted with Airport Network Solutions, which said back in 2003 that it would cost $110,000 to add service. I noted in Aug. 2003 that without aggregation and resale they'd never recoup even the modest cost based on their assumptions of users and what they were charging for a day pass ($6.95). The airport apparently bore the cost of installation repaid out of fees rather than requiring its contractor to eat Capx, which is quite odd.
ICOA bought the installing company in Nov. 2003, and it has resale relationships with many aggregators and operates service at quite a wide array of medium-sized airports in North America.
The Sacramento Bee story today says that the contract with ICOA expired and the service never produced revenue sharing that exceeded the cost of the equipment. But the airport now owns the equipment, it appears, and is offering free service.
I like John Dvorak very much: He is a completely entertaining fellow, and I've had a reasonable amount of correspondence with him in which he is charming, bizarre, and hilarious. He takes up causes. He makes people mad. He writes crazy, contrarian things. And today, he decided to be wrong, wrong, wrong! about metro-scale Wi-Fi.
Read his column first and let me dissect it for you.
"...there is no way that any cash-strapped city -- a category that appears to comprise all of them -- will not succumb to the financial benefit of pulling the plug on this free service, if it's ever implemented in the first place.": Except that outside of St. Cloud, Florida, virtually all free service is being provided by commercial firms who have committed to offering a little or extensive free service on their dime, not the city's.
"Even restaurants, coffee shops and airports that have free Wi-Fi do it only as an inducement to keep people in their facilities. And often those initiatives are undone by a slick salesperson who can show the business how to "monetize" their Wi-Fi.": Now, I agree with the first part, not the second. Wi-Fi was, at one point, a carrot to bring in customers, especially at off times. That's backfired in some places, where they have table squatters using laptops all day for one cup (or no cup) of coffee. (See my story on Victrola and their Wi-Fi free weekends from a year ago.)
But if everyone offers free Wi-Fi, how do you charge for it? You can't. Clusters of coffeeshops now all offer Wi-Fi for free rather than some offering it and some not, or some charging. In Seattle, an early Wi-Fi and coffeeshop culture mecca, I've seen many many cafes switch to free Wi-Fi because they couldn't get paying customers--and paying customers harass baristas about service being unavailable, too.
"In situations such as the various JetBlue terminals around the country where fliers can log on for free, the airports themselves, under pressure from the pay-as-you-go services, are telling the carrier to stop it.": Bzzzt, thanks for playing. That's only happening in Boston and not to JetBlue. And the FCC has to rule on whether it's very clear ruling a while ago that landlords can't restrict the use of unlicensed spectrum applies to Boston-Logan in a dispute with one airline over free Wi-Fi.
"Vendor after vendor will lobby city governments to take over the system and do a revenue split: 'Look, it's easier if we do it. You have to do no work, and we'll give you half the profit.'": That logic would work only if municipalities were, in general, paying for their Wi-Fi networks. They are not. In all of the major and most of the minor metro-scale networks now built, being built, or out for bid or RFI, vendors will bear 100 percent of the cost and will be required by contract to offer a minimum amount of free access. MetroFi, in its successful Portland, Ore., bid, has both free (with ads) and fee (no ads) service.
Dvorak ignores the cost conservation that is a huge part of metro-scale wireless, whether Wi-Fi, WiMax, 4.9 GHz proprietary, or what have you. There is a lot of money spent by cities that will be reduced or recovered when they can pay a much smaller amount to utilize a vendor-owned and -operated metro-scale network. For instance, Phila. will convert hundreds of leased digital lines with high recurring monthly fees to faster WiMax-like service with EarthLink. Phila. might save millions of dollars a year with little to no CapX (born by EarthLink) when the network goes live. And that's independent of whether residential city-wide Wi-Fi takes off.
Some cities, prior to Phila.'s 2004 signal that they'd put out a city-wide bid, had already built Wi-Fi-based public safety networks that were meant, in part, to replace aging analog equipment and celluar CDPD networks. That's where the money is: cost conservation equals just as much money in the bank as new revenue. It shrinks government, it reduces costs, and it produces more efficiency.
"And I can't imagine how much money will be thrown into a public-relations campaign against free Wi-Fi ("It causes cancer!") when someone figures out that it could easily result in universal free phone service using various voice-over-IP subsystems. Mum's the word on that. Shhhhh!": John, would you please nudge Dick Cheney in your undisclosed location and ask him for a newspaper? Voice over IP is a foundation of the interest by cities and citizens in many of these metro-scale networks.
"...there are few municipal officials who understand any of this technology, and none can withstand any orchestrated attack on the idea. It's politically risky when their support would have to come from a public that is so easily manipulated by corporate public-relations propaganda.": The fact is that the risk was in using taxpayer dollars (so called, even though not all city revenue comes from taxpayer dollars...). With that risk offset, and with private companies now funding and owning the networks they build, there's hardly cause for this prediction.
Most metro-scale Wi-Fi and wireless will be charged at a fee, MetroFi and Google's low-speed free offering in San Francisco notwithstanding. For eventual broadband levels of service everywhere and for most service in most places, people will be paying a monthly fee for access. Metro-scale free Wi-Fi won't go away; it will continue to be limited, useful, and advantageous, but not ubiquitous.
If John's contrarian, does this make me a counter-insurgent?
The company that has unwired several cities in the South Bay wins the contract to build service across Portland, Ore.: I grew up in Eugene, and let me just tell you that Portland is completely unlike Sunnyvale, Cupertino, and Santa Clara, MetroFi's three working cities in which you can choose among free, advertising-supported service or $20 per month ad-free service. Portland's politics are complicated, its terrain is topographically diverse, and there's water, water everywhere--one river borders its north, while another runs north/south down its middle, the only major river in North America one of the few in North America that flows north (as opposed to the map direction of "down"; see comments below). Here's the coverage from the local paper, The Oregonian.
Portland is 134 square miles with 540,000 residents; Philadelphia is almost identical in area with three times the population.
The service will include not just citizen access, but municipal services, such as smart parking meter operation and hooking up field employees. Portland has been using meter stations for some time--pay at a central station on a block and stick a bar-coded receipt on your window--so Wi-Fi will probably be a large cost savings over the cellular system they use now.
Portland was a hard-fought contract, with EarthLink Networks and others competing for the prize. While MetroFi has added a city in Illinois and is active in other RFPs, this is a very large win for them for expanding their coverage area.
MetroFi's free with ads option will diffuse the "stealing open Wi-Fi" ethical battle discussed in today's Oregonian. If it's free everywhere (with ads), there's no need to steal a private individual's network connection. It also brings up the issue of whether hotels and other venues that charge for Internet access in Portland will see a large decrease in revenue.
iPass added the Cloud, Swisscom, to Cross 50,000 hotspot boundary: It's not clear how many hotspots there are in the world above the somewhat formally counted and enumerated 100,000 that JiWire has in their database. But it's clear that iPass having 50,000 locations in its aggregated network that it resells to corporate clients gives it about half of all the professionally operated, purposely public and persistent hotspots on the planet.
The Cloud has been aggressively reselling its network to others, and their 6,000 locations are scattered across the UK. While it started in pubs and cafes, they now cover hotels, conference centers, and airports, and have nine city-centre/center projects going life in the next several months. Swisscom has 1,000 locations in Switzerland, including major rail stations, airports, and hotels.
iPass charges per minute fees (often with maximum per-day caps) for access via dial-up, wired connections, and Wi-Fi. They aggregate usage across a company so that many traveling users don't each need their own account, but can use corporate logins and corporate security policies for their data and their connections while they travel.
Technology Review says cell phones on a plane aren't avionics risk: The author says that frequencies which cell phones would interact with avionics are for unimportant systems. But this disregards a key point in the IEEE Spectrum article that he's in part trying to minimize the importance of: that avionics equipment is designed with much lower tolerance for interference than the personal electronics and computer gear that might be used. The electronics we carry with us have much broader legal limits for out-of-band emissions, which can range far away from the actual frequencies a device is designed to use. Avionics on distant bands might be overwhelmed because it's not designed to cope with that much interference.
There's just no way to know without further study, and I don't see why further study is a bad idea in this case. Testing 100 cell phones on a plane of many makes, models, and ages while monitoring a range of avionics equipment (with and without a picocell on board) would be a great start. I'm willing to wait; what's the rush in getting cell phones used on board, anyway?
Further, this article states in an odd way that Wi-Fi costs $500,000 to install on a plane while a picocell will cost $100,000. Not really. What the author seems to have meant to say is that Boeing's satellite-based Internet access system that uses Wi-Fi for distribution on a plane costs $500,000. In talking to several firms that hope to win the air-to-ground spectrum auction that the FCC is running in early May, the cost per plane will be much closer to the lower number than the higher one. The weight of equipment and complexity of gear is much lower, and retrofitting is thus faster and easier, too.
Wi-Fi Networking News has received a nomination for The Webby Awards: The site is one of five finalists in the telecommunications category. The Webbies have an interesting history, having come out of the highly financed dotcom days, dimmed a little during the crash, and then re-emerged as an interesting and separate force, part of a group that's promoting the notion of good design, good technology, and good content. If you look at the membership, it's a signal honor that judges chose this site among the finalists.
The Webby Awards have both a judged and a vox populi component: You can vote for my fine site, if you believe it's the best of the five in its category, by visiting the People's Voice Webby Awards site.
Thanks for the support of readers! I do this for you, honestly. This is a business venture, certainly (although it didn't start as one), but it's the fun of writing, of talking to people, of helping advance knowledge that keeps me blogging daily.
It's easier than running a nuclear reactor--"That's noo-cyoo-ler, son"--but harder than it should be: A Geek Factor column I wrote for Macworld magazine deals with setting up a home hotspot with public and private parts. I took as an assumption that a virtual private network (VPN) connection would be a hassle for an average human being and that you would want to keep your own network traffic completely private. While written for a Mac magazine, it works for any network.
My approach requires two Wi-Fi routers: an inside, WPA-protected gateway for internal traffic that sends its outbound data over its WAN port to a LAN port on an outside, unprotected gateway, which is itself connected to the Internet DSL/cable modem. I chose a Buffalo router model that has privacy separation included as an option. That option can be disabled on the protected inside router, and enabled on the outside router.
Privacy separation should protect the packets that are passing from the inside users over the outside router's Ethernet connection; without this feature, it's possible that a Wi-Fi-connected user to the public hotspot side could sniff passing Ethernet traffic. That's not true on every router because Ethernet and Wi-Fi are bridged, which might reduce promiscuity of traffic, but there's no way to tell explicitly without a switch such as Buffalo uses.
Note to Buffalo: A little more documentation on Privacy Separator would be useful. While it's mentioned in press releases and documented as a switch in manuals, there's no detailed explanation of what it does easily available.
MetroFi may unwire Foster City: This town of under 30,000 south of San Mateo considers an offer from MetroFi for citywide service in exchange for a franchise fee. MetroFi offers service in cities in the South Bay.
Pittsburgh poised to sign with US Wireless: EarthLink's bid for Pittsburgh was rejected partly because their goal is city wide while the City Council wants a quick downtown network erected by July.
Grand Rapids gets 20 bid intent letters: The Michigan town did some extensive auditioning of potential dance partners for a network covering its 45 square miles. Now 20 firms appear ready to bid. This puts the town in the catbird seat in a way that cities don't usually find themselves vis-a-vis technology partners.
This story shows a significant increase of thefts, but a small number overall: San Francisco saw 48 reported laptop thefts in 2005 over 18 in 2004, although this can certainly relate to the increase in laptop usage in public spaces. Eighteen thefts were reported so far in 2006. One victim was stabbed Mar. 16 when he simply tried to stand up when an accomplice of a man leaning towards him made a grab for his computer; although he recovered. Laptops are sold for $200 to $300 on the street.
Rather than try to stop a theft, try to prevent it. One lock expert told me the only security cable he uses comes from PC Guardian--he opts for their combination lock, which he was unable to pick. (He taught me by phone how to pick combination locks offered by other manufacturers, which took me about 45 seconds the first time I tried it using an inobtrusive pick.)
You can also opt for tracking software. CyberAngel Security offers a package that can protect your data transmits an alert. They partnered with Skyhook so that a Wi-Fi connection will allow the laptop software to send an approximate location using nearby Wi-Fi receivers as both locators and conduits.
Slappingturtle.com has just released a beta for Mac oS X users that can turn a MacBook Pro (the new Intel-based model) into an alarm system using the built-in iSight camera, infrared remote control, and other tools. Ultimately, they hope to let the system upload a photo snapped when security is violated.
Did MIMO run over Eileen McCluskey's bike as a kid? I'm slightly baffled by the treatment of MIMO technology in this otherwise reasonable set of consumer advice about Wi-Fi in the Boston Globe yesterday. It's possible the single source cited in the article poisoned her information well.
"These Multi Input, Multi Output gadgets achieve excellent signal quality and range by hogging the wireless spectrum up to 219 yards away. If you live in the city or suburbs, your MIMO router will knock out your wireless-enabled neighbors' connections." Huh?
Here's my letter to the editor just sent off to the Globe:
"Less Is More" (Apr. 9, 2005) contains a glaring error regarding multiple antenna wireless networking. The reporter says that MIMO (multiple not "multi" input/output) gateways hog spectrum and knock out neighbors' reception. This is entirely untrue. MIMO gateways for Wi-Fi, unlike previous range-extending Wi-Fi, are more sensitive receivers not more powerful transmitters.
It used to be that to extend range, you pumped up the volume (increased signal power output), which could interfere with neighboring networks. MIMO uses several antennas to better reconcile radio signals as they reflect off walls and metal objects. This allows the technology to more discretely receive fainter or less clear signals from further away.
There is the danger that a newer form of Wi-Fi, called 802.11n, that will be on the market soon may interfere with older networks in some cases, but Wi-Fi product makers haven't finalized the standard, and it's one of their key concerns.
The other also states that MIMO won't work with free Wi-Fi hotspots. This is entirely untrue. Current MIMO gateways and future gatweays that use MIMO as one piece in a faster Wi-Fi standard are entirely compatible with Wi-Fi equipment sold as long as ago as the first devices in 1999. MIMO devices won't communicate at their highest possible speeds except with compatible devices from the same maker (today) or with the newer standard with any maker (in about 3 to 6 months).
EarthLink and Google won the right to negotiate a contract to provide Wi-Fi service to San Francisco, but privacy advocates are out in force: Why is this an issue here and not in Philadelphia? Because EarthLink is solo in Philadelphia--so far--and hasn't discussed privacy implications much. Google, on the other hand, has talked very specifically about how they'll track and store data about individual behavior to feed them ads. They've even applied for patents for that kind of tracking and delivery.
Interestingly, these kinds of concerns have often been brushed aside or marginalized. This time, with a high-flyer like Google involved, mainstream media and privacy-oriented Web sites and organizations are covering it quite extensively.
The San Francisco Chronicle brings up how requiring a login identity to use the free Google service would allow extremely fine tracking of a person's whereabouts or path through a city even for those using pseudonyms. Google said in its bid, the Chronicle reports, that tracking data would be stored for up to 180 days. The Electronic Frontier Foundation (EFF) points out here that once this data exists, law enforcement would come looking to use it. The SF technology department's head Chris Vein, said privacy was an important factor in evaluating bids, but it's hard to see how that's the case yet--the city hasn't started its negotiations, and thus it's unclear whether they'll push for more protection than is in the bid. The SF Metro Connect proposal (Seakay, Cisco, and IBM) had strong privacy guarantees.
The AP notes that Google won't comment on the privacy concerns.
The Media Alliance evaluated the six bids that San Francisco considered, producing a chart that evaluates what's being offered for protecting privacy and what's on the table for digital inclusiveness. For instance, they note that the Philadelphia plan EarthLink has agreed to include advance payments and other methods of providing computers and access to lower-income residents. Their SF bid has no such provisions, although, again, this might happen during negotiations of the details.
EFF and the Electronic Privacy Information Center (EPIC) assembled this comprehensive comparison of privacy and information gathering details among the six bids.
Jeff Chester raised a number of these issues in his The Nation article two weeks ago, by the way, which I covered at the time, and which received less attention as the EarthLink/Google bid was still one of six.
Meanwhile, Google and EarthLink might work together on another city notwithstanding Google's statement a few days ago that they plan no more Wi-Fi networks beyond Mountain View and San Francisco.
Pardon me as I pick my jaw up, but Freescale has withdrawn from the UWB Forum: Freescale's predecessor company pioneered ultrawideband (UWB); it was bought by Motorola and its efforts spun off as part of Freescale. Both Freescale and Motorola were key members of the UWB Forum, which represented the opposing camp to what became the WiMedia Alliance. The two groups contended in the IEEE 802.15.3a task group deciding on high-speed, short-range physical layer networking technology. Neither budged, and Freescale used procedural and voting methods to keep a 75-percent majority from being reached. The IEEE task group voted to disband in January.
The UWB Forum has been the only bulwark of direct sequence UWB, Freescale's original and ongoing flavor, against the multi-band OFDM approach that Intel and many, many other major semiconductor, consumer electronic, and computer equipment manufacturers have put their efforts behind.
Thus I am shocked--but not the jaded and veteran EE Times reporter to whose story I link--that Freescale and Motorola have decided to exit the UWB Forum to singlemindedly pursue Cable Free USB, their flavor of USB 2.0 without wires. Apparently, participating in a group focused on "regulations, certification, interoperability and common signaling [schemes]," as Freescale UWB head Martin Rofheart is quoted as saying, splits their focus on pushing product out the door.
Freescale's Cable Free USB is not backed by the USB Implementors Forum, which controls the specification through certification, trademarks, and testing. They state, "Certified Wireless USB will support robust high-speed wireless connectivity by utilizing the common WiMedia MB-OFDM Ultra-wideband (UWB) radio platform as developed by the WiMedia Alliance." A quick look at the companies represented on the WiMedia and USB-IF boards makes it pretty clear: Intel, HP, Microsoft, and Philips are on both boards.
The WiMedia Alliance also scored what some see as a large victory recently by having the Bluetooth SIG retreat from its previous position in supporting both forms of UWB to backing just the WiMedia's flavor due to what the group described as a lack of interest in the Freescale flavor.
Interesting development that open-source Wi-Fi gateway firmware project incorporates Sputnik software: The Sputnik Agent allows a gateway to be managed by Sputnik's central aggregation software. The DD-WRT project has gained props both from its extensive installed base in Germany, as well as Fon using it as the base on which they will develop their grassroots hotspot company.
Google and EarthLink were identified as the top candidate to build San Francisco's citywide network at their own expense: This allows the city to start negotiating with the two firms on details. Beware timetables. Philadelphia and EarthLink have spent months negotiating, and then the mayor's office apparently dumped hundreds of pages of detail about five city contracts and a sixth private one on city council members' desks hoping for a rubber stamp. Those contracts are not yet approved by the council.
The AP reports that Google is clearly stating they have no plans beyond SF and Mountain View, which has over 300 access points installed already and is in testing mode. Google's Chris Sacca, one of the municipal networking managers, said at David Isenberg's Freedom to Connect (F2C) conference this week that Google will act like a tenant on EarthLink's network, buying access from them to give away at 300 Kbps, according to Esme Vos.
I filed this story for Secure Enterprise about affordable, solid options for WPA Enterprise and 802.1X authentication for SMBs (small-to-medium-sized businesses): The article, which ran last month (sorry for the late ego link, folks), took as a prerequisite that I wanted to review products that required little to no network administrative knowledge. They should be as turnkey as possible, cost $1,000 or less for 25 seats, and offer standard PEAPv0 with MSCHAPv2 authentication to avoid requiring installing a client under Windows XP SP2.
The results were quite lovely. Several companies were invited, and of those that responded and met the criteria, I found four good solutions each for particular niches. Two outsourced 802.1X vendors, WiTopia.net and Boxedwireless, offer good, solid, and simple options that work reliably. Pricing is cheaper at WiTopia, but Boxedwireless offers EAP-TLS (authentication via individual certificates), which is otherwise expensive and tricky to set up even for a large business. The two hosted server solutions were also just dandy. Elektron is flexible, not overwhelming to learn, and relatively inexpensive. Radiator has more options than you can shake a stick out and isn't for the faint of heart, but it's incredibly powerful. It's too sophisticated for most SMBs, but it also can handle the most obscure and particular aspects of Wi-Fi authentication coupled with RADIUS and AAA granularity.
Note that LucidLink went out of business before our invitation letters went out, and that McAfee declined to participate because it hadn't retooled an offering for the SMB market yet. Funk and Meetinghouse were too expensive, but that per-seat expenses decreases for somewhat larger operations (say 100 to 200 users), and they offer extensive support options not provided by any of the firms featured in this article.
Several authorities in Las Vegas Valley consider Wi-Fi: Sprint has opted to unwire Henderson, while Clark County and Las Vegas consider a joint proposal for the entire valley. The first RFP covers feasibility. Sprint and Cox are both interested. North Las Vegas is interested, but won't be part of the first RFP.
Trial networks are set up for eventual state-wide coverage: The Rhode Island Wireless Innovation Networks (RI-WINS) has been discussing state-wide wireless access for so long that when they began is seemed ludicrous; now, merely difficult. The idea now puts WiMax in the core and Wi-Fi at the edge for access. The state comprises 1,000 square miles, and the network could be running within two years. The network would serve private, public, and residential purposes. The group estimates 120 WiMax base stations would provide core access, although 240 WiMax base stations are planned for Philadelphia's 135 square miles.
NetGear announced this morning shipments of its draft-N based devices; D-Link says later this month: The two firms will be the earliest out of the gate with consumer devices based on the draft standard of 802.11n, which will offer raw rates of 200 to 600 Mbps, with net throughput at a minimum of well over 100 Mbps. NetGear has a gigabit Ethernet line at higher cost, while D-Link has only 100 Mbps Ethernet at the top end. NetGear promises 300 Mpbs of raw speed; D-Link 100 Mbps of net speed.
A majority railway authority in the Bay Area and counties has issued a request for companies to test broadband wireless along one route: The Capitol Corridor Joint Powers Authority (CCJPA) operates along a 171-mile route that's packed with commuters--1.25 million passenger trips a year. Folks commute in from the foothills and mountains, from the SF Bay to the state capital, and among all points in between. Sixteen stations serve the route from Auburn (northeast of Sacramento) to San Jose. The CCJPA operates this trans-county service with BART providing management support under contract.
It's an ideal place for commuters to obtain broadband service on trains, and it's ideal for the authority to use broadband for a huge variety of operational necessities, including on-board security, telemetry, train positioning, and passenger services. The CC line has a few trains hooked up with satellite-based Wi-Fi, and has for some time, but they looking for a much more comprehensive solution that leverages all the benefits of the line's length.
The RFI asks interested companies to consider funding their own trials from Aug. 2006 to Feb. 2007, the results of which will be studied closely not just by the CCJPA, but also by a host of other involved agencies, including Amtrak, the Altamount Commuter Express (ACE), Metrolink down in Los Angeles county, and Caltrans (the state department of transportation). ACE has the longest-running on-board Internet access service--free, even--for its long commuter route from Stockton to San Jose; they just upgraded its service.
The project's principal planner, Jim Allison, said in an interview this morning, "The timing is right to test this sort of application, to learn from these tests. There should be a viable market for these systems in both the urban and the rural settings." Allison works on CCJPA projects under BART's management contract with the authority.
Allison said that no backhaul method would be ruled out, although it's clear from examining the RFI that WiMax might wind up the best technology because of several factors. The CCJPA is explicitly allowing consideration in this RFI for using railway property, rights of way, and fiber optic (where available) to provide service to nearby businesses and residences, as well for trains. User access must be via 802.11b/g with a minimum of 3 Mbps for railway operations (1.5 Mbps in each direction), and 750 Kbps/250 Kbps minimum for passengers.
Because this isn't an RFP (request for proposals), companies can experiment without prejudice, and participation in the RFI process isn't necessary to respond later to an RFP . But the RFI will inform the procurement and RFP process, and it may be that lessons learned here could roll out in a very similar fashion to railways across the region or state. One to three proposals will be accepted for trials along the route, which encompasses hills, mountains, urban cores, and rural populations. An information meeting is April 12, with a submission deadline of May 23. A selection or selections will be announced July 3.
Fiber may be of particular interest as some railways own their own, while others have private fiber on their right of way that can be leased. Pumping a local wireless connection right into fiber could enable one set of services at train stations or other railroad property, as well as providing significant bandwidth for licensed wireless relays to distribute service out across a town. (This is, in fact, part of EarthLink's municipal broadband model in which fiber is used at very few points with licensed wireless used to aggregate traffic to fiber points of presence from clusters of WiMax-like Canopy base stations.)
The CCJPA expects this trial to attract worldwide attention because of the interest by almost all rail authorities in adding broadband. "There's a lot of rail agencies in California and frankly around the states and internationally that would be interested in this as well," Allison said.
The passengers on this commuter rail are fairly likely to use a Wi-Fi network if one might judge demographically. The CCJPA found in a 2005 study that average income was $80,000 and 48th percentile (near median) income was $75,000. 56% of passengers travel for business, and 71% of that group carry a laptop or PDA. No word on how many of the 44% carry laptops--students are a likely large demographic, too. Even better, 33% of computer and PDA users have connected to the existing Wi-Fi service on board for an average of 78 minutes. The route passes by frequent convention sites in San Jose and Santa Clara (the Great America stop).
Of course, one consideration in ultimately building this network is increasing ridership, a perennial concern on most public transportation systems. If operational costs can be conserved through efficiency or lower expenses in using a privately built network, and if passengers become more productive (or have more enjoyment) and thus ride the rails more, that means lower costs per passenger (lower fixed costs) and higher passenger revenue. Allison noted, "If we're rolling along with empty seats, you put another person in that seat--that's instant revenue."
(A personal note here. My father once owned the Fremont Centerville train station during an odd period in its life when it was a furniture store. It's been a train station again for several years, and is part of the ACE system. It's a nice turn of events when my family's past catches up with my present interests.)
T-Mobile pays small fortune to gain three L.A. airports: The Los Angeles terminal has frustrated high-tech travelers for years as other major airports (Chicago aside) gained their Wi-Fi access. With Chicago finally coming online in both airports, LAX now has a plan. T-Mobile will add access there and at the Ontario and Van Nuys airport. The operating authority's press release notes that T-Mobile will pay at least $4.8 million per year for access to customers of these three airports plus 25 cents per session on top of that minimum fee. They've also committed to $1m per year in capital expense.
Los Angeles International sees 61.5m passengers through its doors each year, while Ontario has 7m a year. Van Nuys is dedicated to general aviation, which is everything outside of regularly scheduled commercial flights, with 500,000 takeoffs and landings each year. T-Mobile needs to average about a dime per person at these airports in revenue to break even on their payments.
Unfortunately, T-Mobile has few relationships with other hotspot aggregators--only with corporate-scale aggregators and international roaming partners--so LAX, San Francisco, and other T-Mobile airports will remain expensive for those who aren't T-Mobile HotSpot subscribers.
Five complete bids received for May 10 FCC auction of 4 MHz in the 800 MHz band: This heavily awaited auction will provide one or two segments of high-quality air-to-ground spectrum for a lucky bidder or two. Twelve bidders have so far submitted their applications, five of which are complete. Boeing (in the form of Connexion by Boeing) will sit this auction out, the Chicago Tribune reports. (Download the public notice and appendixes with bidders here.)
It is somewhat staggering. Boeing played its cards close to the vest for the last year while the auction details reached final form, and it's likely that having run the numbers for this enterprise, they decided they couldn't make money at what they anticipate the bidding would go to. They certainly have the technology and the ability to purchase the technology to make the spectrum work and to integrate with existing and new ground operations.
Boeing may have opted to wait out the process to reduce their capital expense in running an additional network, as it almost 100% certain that whatever winning bidders build a domestic broadband in-flight service will want to have strong ties to Boeing for customer handoff. Some aircraft will likely contain both systems. The weight of the air-to-ground antenna and components is expected to be less than an infant.
Boeing's only satellite competitor, OnAir, didn't submit a bid but was hardly expected to. OnAir hasn't yet launched its full-on offering which relies on Inmarsat's fourth-generation BGAN satellites that offer pinpointed broadband. (Coincidentally, OnAir did put out this press release today noting that Air France will receive the first Airbus equipped with OnAir GSM equipment for in-air cell calling in early 2007. Air France will use the plane for a six-month trial. OnAir previously said two other European airlines plan to launch trials later this year with retrofitted planes. OnAir is a joint venture of Airbus and airline systems integrator SITA with the former assets of Tenzing. Unlike domestic U.S. potential broadband operators, OnAir plans to focus first on GSM service and later on broadband.)
The seven incomplete bids including proposals from AirCell, which already operates a national ground-station network for general (non-commercial) aviation, and is expected to qualify for bidding credits as a smaller business, and Verizon Airfone, which operates the only in-flight phone service that remained after competitors fled the market. Airfone will continue to operate its current service for up to two years following the auction's award. JetBlue's LiveTV division also has an incomplete bid.
Incomplete bidders have until April 17 to rectify the missing components in their applications. That's also the date for a deposit of the required upfront payment of $100,000 for bidders who haven't previously defaulted on payments; $150,000 for that category of bidder.
The spectrum up for auction is 4 MHz total proposed for auction in one of three potential configurations: 3 MHz/1 MHz, 1 MHz/3 MHz, and 3 MHz/3 MHz overlapping. The 3 MHz would be 1.5 MHz in two separate parts of the 800 MHz band to provide separation for uplink and downlink. It's expected that at least a 1:1 MHz:Mbps ratio will work, which makes the 3 MHz options the only reasonable outcome. At least one firm, AirCell, would use EVDO Rev. A for its linkage, which would allow it to use commodity equipment which operates at a rate somewhat above the expected performance for this band.
Five complete bids were received, four of which were from firms of the smallest category which receive a 25% bidding credit. I didn't recognize any of the five firms' names, some of which are obviously consortiums put together for bidding purposes. Each of these five firms have bid on all six licenses, which is likely strategy for all bidders.
My guess at the auction's value? I have zero insider knowledge, and I estimate $40 to $60 million for the winning configuration with Verizon Airfone almost certain to be the winning bidder. If the domestic airline industry were stronger, I'd suspect this could auction could have netted the government $100 to $150 million.
Go Networks emerged from stealth at the cell industry trade show today, announcing their line of cellular-style Wi-Fi products for metro-scale outdoor deployment: In a briefing last week, the company explained that they use a combination of multiple radios and multiple antennas to provide broader areas of coverage than any mesh nodes on the market today. Their approach uses beam forming, originally pioneered as beam steering by defunct Vivato, which re-emerged as a technology in the era of multiple-antenna (MIMO) Wi-Fi. With beam forming, less power can be used to reach greater distances through redundant, targeted paths. Receive sensitivity is higher as well. (Full-on MIMO can send different data streams over divergent paths, reusing spectrum and increasing throughput at the expense of redundancy.)
The company calls its beam-steering technology xRF and its cellular routing system xCell. The firm uses the 802.11a (5 GHz band) for backhaul, and has base station models that sport multiple 802.11b/g radios. xCell allows them to segregate end users by traffic type, service level, or protocol (b versus g) in much the same way that some enterprise switched networks offer.
This doesn't require changes to the Wi-Fi adapter; rather, a combination of virtual and real SSIDs and built-in Wi-Fi protocol features let them control which network given devices connect to and even lets them force adapters to migrate to other networks. "Whatever devices you will have, whether it's a laptop or a handheld device, if it supports 802.11 or later on 802.16e, it will work with our equipment and you will get all the benefits," said Yuval Mor, director of marketing for the firm, and formerly with Cisco in their high-end routing group
Mor said that Go's approach will require 50 percent fewer nodes per square mile than competing solutions because of its antenna and routing approach, reducing capital expense and operating expense. Because each node can have multiple Wi-Fi radios, more service can be provided in any region covered by a single node. For those counting, Tropos's estimates tend to be 20 to 30 per square mile divided into clusters of about five, each cluster sporting a separate 900 MHz or 5 GHz uplink to a backbone network.
There are two types of outdoor nodes in Go's system: a micro base station and a pico base station. The micro base station features one backhaul radio and two Wi-Fi radios. The pico base station, intended as a niche space filler for underserved areas, has two backhaul radios to allow meshing and one Wi-Fi radio. The micro station can mesh as well, but Go is using mesh as a supplement to their approach. Mesh routing always occurs on the 5 GHz radios, as with BelAir, SkyPilot, and Strix equipment.
The micro base station uses 120-degree sectorized antennas to qualify under higher point-to-point signal limits in an FCC rule originally granted for Vivato. The pico station has an omnidirectional antenna. An indoor receiver is also available; it's designed to better relay service in from outdoor nodes. A back-office device aggregates control and handles xCell configuration among other features.
Mor said that the company is looking away from simple best-effort network access and into new markets that are developing "where you need a lot more guaranteed quality of service," such as voice over IP calling, video, and location-based services.
The company has had beta tests underway for several months in Asia and Europe, said. Mor said that the company views China as one of the largest potential growth markets, but will be actively competing in US and Europe as well.
I've been waiting for the first mesh/metro firm to deploy a beam forming approach with multiple antennas, because MIMO is clearly an effective technology in difficult RF environments. Although MIMO works best in areas with reflection, dense outdoor spaces should provide the multipath reflection that gives a "workaround" to areas of interference or obstacles that can't be penetrated with a direct beam.
Posted by Glenn Fleishman at 12:19 PM | Permanent Link | Categories:
Tempe closes gaps: NeoReach is reportedly working to finish a few gaps in their coverage. They deployed 600 Strix Systems access points, and had hoped to have full coverage by the end of February. They'll be installing 50 more locations, and the city won't penalize them for missing the deadline as they are close to the effective requirement. This shows that even a relatively flat terrain can introduced unexpected difficulties.
Sacramento starts on Wi-Fi network: Strix Systems gear will also be used to build out 10 square miles of this California capitol with over 2,000 access points planned. NeoReach's parent firm MobilePro is behind this deployment, too. The ribbon-cutting is tomorrow at which event various technologies will be demonstrated.
Scranton, Wilkes-Barre move closer to downtown-Fi: The University of Scranton is a month into a pilot program for some local Wi-Fi around a courthouse. Wilkes-Barre has a coalition of partners working with a partner to build its service out.
A bill in progress in New Hampshire would allow municipal infrastructure support for broadband: Towns that lack Internet connections could sell bonds to provide service. This is the first instance of this direction I've seen, but it makes sense in the Live Free or Die state.
The Washington Post rounds up local hotspots, the Journal covers travel tips and Wi-Fi woes (sub. req.): The Washington Post looks into D.C. and environs hotspots, providing a detailed review of the quality of the location, its food, and other notes. These types of stories have become less common as Wi-Fi turns into a component expected without comment, like an appropriate nitrogen/oxygen mix in a cafe's air or tap water on demand.
The Journal looks briefly into what makes one hotspot reliable and another not--I disagree with their conclusion that it's free versus fee, of course. And Jim Carlton files a column on traveling tips, quoting me at the end, and a few colleagues. My friend and colleague Adam Engst apparently carries a PowerSquid with him, which could provoke a number of ribald comments, but let's just say, a PowerSquid is a good way to make friends in an airport.
The corporate aggregator and end-point security provider partners with T-Mobile International across Europe: By the end of second quarter, iPass will add another 8,000 hotspot locations from T-Mobile's European operations. While iPass secured a relationship to resell access to U.S.-based T-Mobile points well over two years ago, the European partnership was more complex, said Rick Bilodeau, an iPass director. "This was a long time coming in terms of negotiations because we had to negotiate and get agreement from all the different T-Mobile subsidiaries and the T-Com subsidiary," he said.
The deal offers metered service up to a daily cap, similar to the T-Mobile USA arrangement. The partnership adds 3,500 hotels, 2,700 food establishments, and a number of Continental and UK airports. The company estimates it will top 60,000 hotspot locations worldwide when these locations are certified and added into their network in a few months. (This includes both wireline-only hotels and primarily Wi-Fi in other locations.)
While T-Mobile International and T-Mobile USA are part of the Wireless Broadband Alliance, a group of hotspot operators worldwide that have negotiated standardized roaming and authentication agreements, iPass isn't an operator, and has had to pursue these relationships separately.
Bilodeau spoke of the next frontier as mobile data, given the remarkably high charges for roaming access to 2.5G and 3G services while outside of a home country area. "We have EVDO in the US as a service that we bundle into the fold: one bill that covers your 3G and Wi-Fi and your dial, if you're still using dial. Our goal is to grow that internationally," he said.
iPass member companies use software provided and customized by the firm to offer a single unified login tied into a corporate authentication system, along with end-point security options that can require a virtual private network, up-to-date anti-virus software definitions, and/or an active firewall to connect to the Internet while outside of the office. iPass aggregates metered charges across a company's employees obviating the need for individual accounts with multiple providers for dial-up, wireline, and Wi-Fi access.