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.
I disagree with CNet's take on Whisher: News.com's Marguerite Reardon paints Whisher as a way for people to share their broadband connections with others with a bit more security than simply leaving a network open. I can't see that has the kind of utility or uptake needed to make Whisher work. Rather, my view is that you can have 100,000 buddy groups with 2 to 10 people each in them, coupled with thousands of hotspots that are intentionally open and free in public places, but decide to use Whisher for the services it adds. It's possible that some home users and others would offer up their protected network connection as an open network via the Whisher system, but that seems much less likely than easing access to a secure network for all their friends. I know that's how I'll use Whisher.
I don't see Fon and Whisher as having similar goals. Whisher can work if 10,000 people adopt it (although it may not be affordable to develop). Fon requires at least large clusters of its network nodes for the utility to be great enough. Put this way, if I only use Whisher to connect at networks run by a dozen friends, it's still highly useful--less management, less fuss, and I get file transfer with those friends when I'm on their network. With Fon, if I go to a new city and can't find a convenient Fon location, all my Linusing--freely sharing my Fon connection--is for naught. These are definitely not the same models or utilities.
The device-centric firm and service launches its release version at the DEMO 07 Conference: Devicescape wants to have each piece of Wi-Fi-equipped gear you carry wear a unique identifier, and use lightweight software to allow those devices to log in to hotspot networks, home networks, and, potentially, enterprise networks. Rather than work to cut deals or build software that lets a given VoIP handset connect to a given network, Devicescape is device and network agnostic, working to build as many connections between devices and Wi-Fi networks as possible. I wrote a long article on their goals last year.
Their 1.0 release supports one Linksys phone, the Nokia 770 Internet Tablet, Windows XP SP2, and Windows Mobile 5. However, they added or are near adding several hotspot networks to their system--these connections don't indicate partnerships, but the technical ability for their supported devices to log in. AT&T FreedomLink is now missing from their list, but T-Mobile is still there. (See comment below.) They're working on getting Wayport and Boingo, among other networks, added.
The AirPort Extreme Base Station with 802.11n is now shipping: The software enabler required to update existing Macs that have 802.11n technology built in can also now be purchased from the Apple Store for $1.99. The enabler is included with the $179 Extreme gateway. All Core 2 Duo and Xeon-based Macs with Wi-Fi can be updated, except a single iMac model, but including Mac Pro desktops that had the AirPort Extreme option added. Apple isn't offering 802.11n options for any older Macs; third-party adapters will be required.
The enabler isn't locked to a particular Mac. David Moody, an Apple vice president, said, "You can install it on the all the Macs in your house." The license on the purchase page is even broader: "The software license for the 802.11n Enabler software allows you to install and use it on all computers under your ownership or control."
AirPort Extreme with N can work in either the 2.4 gigahertz (GHz) band, in which 802.11b (AirPort) and 802.11g (AirPort Extreme) operate, as well as in the 5 GHz band, which is less occupied and has greater available frequencies. While the new Extreme includes Wireless Distribution System (WDS) support for linking base stations wirelessly, and will work interoperably with the older AirPort Extreme and AirPort Express models. However, Moody said that the greater range of 802.11n should obviate the need for WDS connections in the home.
In 2.4 GHz, Apple won't allow 40 MHz "wide" channels that, in the absence of other Wi-Fi network signals, could double throughput. Moody explained that Apple has a huge interest in preserving the functionality of Bluetooth, which has shipped alongside Wi-Fi in most Macs in recent years. "We need to make sure Bluetooth and [802.11]g co-exist perfectly," he said. Allowing 40 MHz wide channels in 2.4 GHz would have severely constrained Bluetooth. Starting with version 1.2 of Bluetooth, that short-range networking standard actively avoids frequencies that are in use by Wi-Fi.
Jai Chulani, senior product manager at Apple, suggested that many users would be better served by preserving a legacy 2.4 GHz network for 802.11b/g devices with an existing base station, and plugging that older base station into an Ethernet port on the new AirPort Extreme, which would then operate to its best advantage in 5 GHz. The AirPort Extreme with N is configured to automatically choose the best channel in 5 GHz, but Chulani said that an advanced settings option would allow 5 GHz channel selection. This could be important, because four of the nine channels in 5 GHz that Apple is offering are restricted to a low-power mode.
Chulani also confirmed that the Apple TV could operate in either 2.4 GHz or 5 GHz bands, but that like the AirPort Extreme, the best mode of operation would be automatically selected, and could be manually overridden.
The AirPort Admin Utility has been updated for the new standard with an overhauled interface that, Chulani said, "has two faces." One features more automatic, sensible choices for users who don't need or want to customize configuration. The other includes even more technical detail than earlier releases. For instance, the separate, free AirPort Client Monitor, used to view connected devices and their signal strength, is now part of the AirPort utility. An update to the AirPort Management Utility for configuring multiple base stations at once will be released in the indefinite future. Mac OS X 10.4.8 or Windows XP is required to configure the new base station.
The pipeline for Apple's 802.11n is just revving up, and Macs sold at retail could need enabling. Purchasers of Macs that don't have the newer software installed will have to pay the $1.99 fee unless they also purchase the AirPort Extreme Base Station. It seems likely that the base station will drive the upgrade to 802.11n. The base station started shipping today, and orders placed at the time of announcement should be in purchasers' hands shortly, Moody said.
A small spate of announcements from remote access firm iPass: The company resells access to 75,000 hotspots worldwide and countless dial-up lines, and has added EVDO Rev. A access and satellite roaming via Inmarsat's BGAN service. EVDO Rev. A reportedly runs at 450 to 800 Kbps downstream and 300 to 400 Kbps upstream; testers have found much higher downstream rates but often much lower upstream rates. iPass also said they will support Windows Vista in the second quarter.
While they don't identify which EVDO provider is which, it's easy to guess that iPass is offering service from both Verizon and Sprint, since there are two networks they offer and two providers of such in the U.S. They call them Network A and Network B, and require separate subscriptions for each network. It's likely that the EVDO Rev. A addition is from Sprint. The new offering costs $60 per month for unlimited use and volume discounts can reduce that further. Adapters are extra. This is one of the few cases in which iPass has a recurring per user fee, and I imagine that if the cell operators ever offer a pay-as-you-go system, iPass will be one of the first to provide it. They were T-Mobile's first roaming partner, too.
Inmarsat hasn't to date offered a simplified access structure for their fourth-generation satellite network known as the Broadband Global Area Network (BGAN). Various companies resell terminals and access, but iPass will have the clearest and most transparent model for a company that may deploy a few terminals and have various employees using the network. BGAN can operate up to 492 Kbps, and charges are levied per megabyte.
Via email, an iPass spokesperson explained that the satellite service will come with two pricing models. A usage-based model will cost $60 per month per user and $7 per megabyte. This can be canceled at any time. More favorable to large corporations is a pooled model which carries a 1-year commitment and must include at least 10 users. The pricing is per user per month with 10 users at 20 MB each costing $120 per month up to 750 MB each for $3,000 per month. Terminals are sold separately and range from $2,000 to $4,500 with most falling in the $2,400 to $2,800 range, iPass said.
(Recall that OnAir and Aeromobile are planning to launch in-flight data services using BGAN eventually--in-flight cell may launch any day now on limited airlines in Europe and Asia--but you can see that the per MB cost on a corporate level makes it impossible for unlimited in-flight satellite-based Internet use. Connexion by Boeing relied on a different set of satellites that carried largely fixed costs, but those costs required millions of sessions a year to produce enough revenue to break even.)
iPass sells mostly to the corporate market where rather than have each roaming employee set up their own accounts with recurring fees, iPass can meter access or provide negotiated monthly rates across an entire organization.
Share your network with friends and colleagues without managing passwords, says Whisher, but that's just for now: Stealth startup Whisher comes out of the woods today with an approach for allowing simple access to secured networks at homes and small businesses, as well as free hotspots. Whisher requires the use of a "heavy" (read: large) client application that's available for Windows, Mac OS X, and Linux. The program handles account and password management; there's no configuration changes or firmware required for access points. "It's not about reflashing routers at all," said Ferran Moreno, Whisher's CEO and co-founder.
To share a network, you create an entry, enter its password (if any), and choose buddies that can access it. "You can create your own Wi-Fi community, by deciding with whom you share and when," Moreno said. Buddies must download and install Whisher, and create accounts. There's no cost for the software or this functionality. Hotspots that want to be open and free can still protect access via WEP and WPA, and use Whisher as the "bar to entry." (The trick is that each copy of Whisher has an encrypted database storing passwords for all open networks, and the passwords for buddy networks.)
The buddy notion bypasses the need to provide friends and colleagues with your network password, or even co-workers in a small-business environment. The centralized management means a single change flows the modified password to your buddies, or to everyone for open networks.
The application includes file transfer software and instant messaging, but both features are currently limited to allow exchanges only over the local Wi-Fi network for registered users. That limit means that someone who installs the Whisher application and doesn't register for an account can still access free hotspots, but if they want additional local features, they'll need to create an account. A hotspot can keep a CD or memory stick handy for customers to install the Whisher software, which doesn't require a restart under Mac OS X, at least, to get up and running.
Whisher represents a pendulum swing back from the Web-based gateway access that I would argue has plagued the hotspot industry. The gateway page took hold before 802.1X supplicants were a viable way to allow access. (Windows XP and Mac OS X have included supplicants for at least three years; Unix/Linux flavors have free and for-fee supplicants available.) If the hotspot concept started up today, it's pretty clear that a standard login methodology around 802.1X would be adopted for simplicity and security. (T-Mobile's Connection Manager uses 802.1X seamlessly, for instance, and has for years.) When you read complaints about how Wi-Fi doesn't work as a public access means, almost always the complaint is either with cost, or with the trouble in producing an authentication session through a gateway page. (I was just wrestling with AT&T FreedomLink's ugly approach to this on a recent trip.)
By putting the intelligence in client software, the company has to deal with supporting multiple platforms, but with the massive heterogeneity of Wi-Fi routers, that may be a simpler win than dealing with creating one's own firmware or pushing out custom routers, which has been Fon's approach.
(Moreno said he worked with Fon founder Martin Varsavsky, but left over differences in achieving similar communities and scale. Varsavsky disputes this over at Business 2.0. Moreno said that Whisher could absolutely work on top of Fon's network as an application to ease access, provide local network services, and even content based on location, as the software could do on any network. CTO and co-founder Mike Puchol has engaged in rather public Fon-baiting and criticism, including a lot of back and forth with Varsavsky in their respective blogs: Varsavsky, Puchol.)
Downloading an application used to be the kiss of death when broadband was scarce, and it sometimes seems antithetical to the Web 2.0 aesthetic of AJAX-based browser intelligence. The Whisher app ties into that, though, by using an embedded browser, and intra-program updates. "It won't be any more welcome page, enter your credentials, and so forth -- it will be a uniform experience," said Moreno, and that's really the key.
The company hopes to see revenue by leveraging its application as a way to deliver advertising for hotspot networks, and as a vehicle for hotspot and metro-scale Wi-Fi operators to resell service through a client that someone may already have installed. As with Devicescape's device-oriented model, Whisher will use Web-based accounts eventually to handle credit card numbers or other payment mechanisms, as well as other centralized account details. "We're leaving it open to be integrated with any WISP or any technology it has for authentication and billing and so on," said Puchol.
Comparing Boingo, Devicescape, Fon, and Whisher
Boingo: Software client for Mac OS X and Windows mediates access to aggregated partner Wi-Fi networks. A monthly subscription fee allows unlimited access via computer to US locations and some international locations. Many international hotspots have metered, negotiated per-minute or other rates. A new VoIP offering includes unlimited VoIP worldwide at participating locations. $22 per month or $8 per session, plus metered rates at some international locations (noted in their hotspot finder). Boingo has north of 60,000 hotspots signed.
Devicescape: Embedded software that Devicescape wants manufacturers to include in their handheld products, and that can be installed on some devices, allows access at public hotspots that require authentication by tunneling login information via DNS. The handheld device stores no authentication information, and uses the tunneled credentials and other details to perform the login; entering credentials and account details occurs at a Web site. Devicescape wants to establish the notion of per-device sub-accounts at hotspots and aggregators, allowing a single user to use more than one phone, game console, camera, and PDA--with perhaps different, small monthly rates for each--at the same time. Free, but requires accounts at hotspot networks. Supported networks include AT&T FreedomLink, Fon, and T-Mobile USA. (Support does not indicate the hotspot operator approves or works with Devicescape's system.)
Fon: Grassroots and commercial network built one at a time, with individuals flashing commercial, commodity gateways from Linksys and others with special firmware, or purchasing typically subsidized routers that are preconfigured. Network sharers, called Foneros, can choose to charge for access to their hotspot or share it for free. Those who charge receive half the revenue. Those who share can access any other shared Fon location for free. Non-Foneros pay for usage everywhere; those who share for free receive none of the revenue from non-members using their location. Fon's model may extend beyond flashed/custom routers soon with the release of Mac OS X software that turns a Mac into a Fon hotspot, allowing routing from a cell data network (cell operators uniformly disallow this in the US, but it would be difficult to track). Fon claims tens of thousands of active Fon hotspots, but I can't find a citation for their current count. Their mapping software shows registered users and active locations.
Whisher: Client application for Windows, Mac OS X, and Linux allows access to password-protected (WEP/WPA) public and private networks coupled with a Web-based account for management. Any Whisher user can access public networks; private networks require inclusion in a buddy list. No special hardware required. Free software, free buddy lists.
Four on San Francisco: Two opinion pieces appeared from parties in favor of San Francisco Wi-Fi, but generally opposed to the method and manner by which it appears it will be deployed. A third supports the current plan. A fourth article outlines the costs and nature of a major fiber roll-out.
Becca Vargo Daggett of the Institute for Local Self-Reliance, an outspoken advocate of municipal ownership of broadband networks, suggests the city's Board of Supervisors reject the EarthLink deal brokered by the mayor. She notes that a recently released independent budget analyst report shows that the city consistently decided against checking on the feasibility of municipal ownership. She thinks the EarthLink/Google deal isn't a good one because of the Wi-Fi bridge requirement, which means that "free" service will still cost something--the cost of a bridge--and that once the deal is in place, there's little chance for competition.
Sasha Magee over at BeyondChron notes that the coverage area and quality that the contract defines are fairly ambiguous, and wonders why the board should approve a document that doesn't spell out the requirements in greater detail. Magee thinks the network will be too slow, provide too little revenue to the city to fund digital divide efforts, and be in place for too long.
At the SF Examiner, Heather Hudson of the University of San Francisco thinks the price tag to the city (free), the potential for competition, and the basic level of service are plenty good enough to just go ahead and build.
Finally, the Examiner also looks into fiber to the premises (FTTP), which would cost $560m and take 15 years to build out in the city, but which Supervisor Tom Ammiano is highly interested in as an a complement to Wi-Fi. Ammiano suggests the city lease use of the network to private providers. He's critical of the Wi-Fi plan because, as others note above, it's unclear how low-income residents will receive it without spending money.
On the other end of the country, Belfast, Maine, gets yet another wireless option: While rural residents have fewer and worse connection options than those in many (but not all) urban areas, it's always interesting to look at exceptions. Belfast is a small city tucked in the part of the coast that's called Down East, and it will have four overlapping wireless data networks. A local provider offer Wi-Fi service across Belfast. An old established provider, Midcoast Internet Solutions, offers broadband wireless already in parts of town (point-to-multipoint); Mainely Wired also covers some of Belfast in a similar fashion; and FairPoint, which is buying Verizon's landlines, will add wireless service--which kind wasn't specified.
Ruckus Wireless adds virtual second network to its metro-oriented Wi-Fi bridges: The MetroFlex DZ offers multiple virtual Wi-Fi network names (SSIDs), which allows a single bridge to communicate at high power with a metro-scale network and at low power to devices in a local network. It also lets the user of the bridge configure strong security on the local virtual network. Ruckus uses MIMO and dynamic power control to focus energy where it needs to go. This reduces interference from and with other devices.
A similar set of devices from rival bridge maker Peplink use traditional high-gain Wi-Fi antennas, and thus cast energy in lots of directions. The MetroFlex DZ has a list price of $149, and uses a 200 milliwatt (mW) radio. Peplink offers two models: the Surf 200BG-AP lists at $189 (200mW) and the Surf 400BG-AP at $289 (400 mW).
digitalculturebooks is looking for nominations for the best technology writing of 2006: The venture between the University of Michigan Press and Library is looking for articles, essays, and blog posts that are "engagingly written for a mass audience," no longer than 5,000 words, and published in 2006. I wrote a million billion words in 2006, and if anyone thinks something I wrote stands out, I won't object to being nominated.
Can I beat the drum any louder about the speed of the wireless LAN in a hotspot? It's hardly been leveraged, despite the long existence of all 802.11g hotspot networks and built-in laptop adapters. With most home broadband running under a couple Mbps, and WLANs in a hotspot able to achieve 20 to 25 Mbps, you'd think that differential could be leveraged for media. Sure, there are issues of fulfillment and security to be dealt with, but there's money in them there WLANs. Stick a media server and a couple terabyte array on the network, and Bob's your uncle.
Starbucks may be considering an in-store media delivery partnership with Apple, based on some tea leaf--excuse me, coffee grounds reading, given that Starbucks now has an area in the iTunes Store, and Starbucks chairman Howard Schultz said they'd offer a digital music fill-up service in their stores within 12 months. Maybe you'd walk up to a kiosk and plug in a USB cable. But it would be more sensible to offer Wi-Fi connections, especially with 802.11n about to flood the marketplace with speeds that could average 100 Mbps, especially a controlled indoor space. (USB nets to about 300 Mbps of throughput.) Wireless USB is also an option if ultrawideband takes off.
In every transportation segment I've spoken to--planes, ferries, trains, buses, and subways--there's great interest in coupling Internet access (which could run 200 Kbps to 2 Mbps, sometimes faster) with on-board media servers feeding streaming media and, potentially, downloadable purchasable media. No one is really taking advantage of the wireless LAN in hotspots or across campuses and cities yet.
Phil Belanger drops a branding bomb: David Haskin reports that Belanger, a founding member of the Wi-Fi Alliance said that FlankSpeed was one of the names suggested for what was ultimately dubbed Wi-Fi. I have to stop and put my head between my knees. Deep breaths. Okay.
Many moons ago, the company that invented Visio--bought by Microsoft for $1.6b in 2000--was searching for names for its product (and ultimately what it renamed the company). The naming consultants looked at the fact that the program could graph and draw, and came up with...Graw. Thank you! Next! The founders of Visio later formed a startup named The Graw Group as a kind of a joke until they left stealth mode as Trumba.
The president of Lakehead University spreads poor information disguised as prudence: In Toronto, at the Wireless Cities Summit, Lakehead president Fred Gilbert repeated the bad science that led him to block wireless networks from being used at his university. On the panel with him was Magda Havas, who is an associate professor at Trent University, and another person who plays fast and loose with microwave studies. Gilbert is quoted as saying, "there are potential health impacts we felt we should be employing a precautionary principle with respect to this technology on our campus." Which is fine if there were a shred of evidence to back that view.
To see how specious Havas's reasoning is, here's her explanation of an earlier problem with microwaves: "When radar was first invented in World War ll we found many radar workers came down with radar sickness, which is what we would now classify as electrosensitivity." Which is totally incorrect. Electrosensitivity, which one study recently showed was non-existent in their testing, has been primarily used to refer to a reaction that some individuals have to electromagnetic fields that contain energy far below the threshold of affecting human tissues or nerves.
Radar sickness--I can't find citations for this precise term related to WWII--could have been the result of exposure to massive amounts of microwave radiation, which is a known problem. In fact, I advise that no one stand near active Wi-Fi or wireless transmitters that use high-gain antennas. These are typically mounted on towers and rooftops, and there's a body of research that shows that at certain thresholds, you can get cellular disruption and long-term health problems. But those levels are well characterized and several orders of magnitude above Wi-Fi and cell phone output. The only study I could find looked at Korean War radar technicians, who had below-average mortality compared to control subjects. (Another reference to radar sickness I find refers to illnesses caused by handling or being near radioactive elements used in Cold War radar installations.)
Havas later compares Wi-Fi use to the use of X-rays to determine a child's shoe fitting as an analogy--the radiation type is vastly different in effect--but it shows her intent to conflate.
Gilbert demonstrated more specious, non-empirical logic when asked by an audience member why, if the studies are all about cell phones, shouldn't the focus be on cell phones? Gilbert responded that "the critical thing is not that there's 100 times difference between the two. The critical thing is the biological effects." Which would mean he believes that there's a magical, perhaps homeopathic property in electromagnetic radiation that affects people regardless of whether the energy passing through someone is below the level necessary to shunt electrons out of their paths.
Havas backs up Gilbert by noting, "There has been very little research on the effects of Wi-Fi because it hasn't been around long enough, so we have to look at technology that is similar to give us the answers as to how concerned we should be about the effects of this technology." Right. And we can extrapolate based on two facts: First, that there are no credible cell phone studies that show long-term or short-term health effects; and second, that Wi-Fi operates at levels far below cell phones, and at greater distances, further reducing any potential effect.
Now before you say--wait, what about that new glioma study that was covered today? Let me stop you. That study, which involved subjects with particular cancers matched against control subjects, and relying on retrospective data (relying on recollection, to boot) can't hold a candle to the other recent study that looked at actual cell phone bills for calling behavior, and had 425,000 subjects with 56,000 using a cell phone for 10 years or more.
As noted in other studies, relying on recollections for sidedness in cell phone use is invariably biased by the subject, suffering from cancer as they are, being more likely to associate the side with the tumor with the side they formerly favored.
Say it ain't so! The 787 sheds wireless media and what seems like Wi-Fi capability in favor of wired connections at each seat. The reconfiguration saves a net weight of 150 pounds, equaling one below-average American sans baggage or one Parisian avec baggage. It's not clear how many wireless systems were involved here, but Boeing already built a standard cabling and conduit system that allows them to tack on per-seat wired connections with a fairly small incremental effort. The wiring will weigh about 50 pounds; the wireless gear toted up to 200 pounds. Boeing says frequency coordination worldwide was the culprit, however.
The folks at Novarum think Wi-Fi might actually work city-wide: Ken Biba and Phil Belanger, the founders of Novarum, released a few statistics from their tests of 14 cities in the US that contain large-scale Wi-Fi networks that are substantially deployed or complete. Their results are the strongest endorsement yet that these sorts of Wi-Fi networks can work, but also provide a better sense about the infrastructure cost required to provide full coverage and broadband speeds. (You can read my interview with Belanger from last November here.)
In a set of emails with Biba and Belanger, I asked based on their testing how many nodes per square mile were truly needed to provide service? Biba explained that the intent of the network mattered, but that about 40 nodes per square mile "plus or minus a few depending on tree cover and terrain" was ideal for suburban installations. This is above the 35 nodes per square mile figure EarthLink was putting out last year, and far above numbers provided as recently as a year ago (in the low to mid-20s). A 100-square-mile network that was budgeted for 25 nodes per square mile that really needs 40 would need an extra 1,500 nodes that could cost $3m to $5m additional.
Belanger's sound bite: "40 is the new 20."
St. Cloud, Florida's free Wi-Fi network was the only one of the 14 with 100-percent service availability, which Novarum defines as not just a signal, but the ability to perform useful work over the connection. The No. 2 slot, Anaheim, Calif., scored just 72 percent, all the way down to 44 percent in Galt, Calif., in position 10. That's the area that needs improvement, and it's a direct correlation to infrastructure density, the Novarum founders say.
Toronto Hydro installed as many as 126 nodes over one square mile, which seems insanely high, except for two factors: first, it's a dense "urban canyon" with a huge number of tall buildings; second, performance was remarkable, with 5 Mbps symmetrical in many locations. This speed is highly competitive with wired broadband, and it's the first time that such consistent performance has been demonstrated in a Wi-Fi network. Other networks might be able to achieve those speeds but throttle per-user bandwidth.
The density of population could allow that many nodes to be profitable, too. Novarum didn't test coverage in higher floors yet, but plan to. Toronto scored only 60 percent on service availability in the tested area, which shows that simply putting up nodes doesn't ensure ubiquity.
Also surprising was how cellular outranked Wi-Fi in the cities surveyed. Eight of the top 10 wireless broadband overall scores went to cell networks--and not a single cell provider but divided among Sprint (three cities), Verizon (three cities), and Cingular (two cities). This measure looked at service availability, performance, ease of use (including login, network discovery, etc.), and value (price versus other factors). Cell can't beat Wi-Fi for speed in ideal cases, but it can for overall experience and ubiquity.
Update: David Haskin offers additional detail in his report on this study.
EarthLink will build Wi-Fi network for the town its headquarters is located in: The detail isn't much, but it's pretty much the same kind of 1 Mbps symmetrical for-fee network they've committed to or have built elsewhere. It's just in their own backyard.
Starting Jan. 26, Vista purchasers can use T-Mobile's US hotspot network at no charge for 90 days: Unlike other deals, the clock starts on a date, not when you purchase Vista or a new computer with Vista installed.
The FCC requires broadband firms to report by Zip code, the UK's regulator Ofcom by actual lines installed: I've come across a situation that's relatively well known in the industry, but I'm not quite sure that the average person gets this. And the average reporter clearly does not. When the FCC says that 99 percent of the US population has access to broadband--not subscribes to, but is passed by service they could choose to subscribe to--they are deriving this in the following specious way: 95 percent of Zip code areas in the US have at least a single subscriber to a broadband service. Further, broadband is 200 Kbps or faster in either direction.
The General Accounting Office released a report in May 2006 critical of this notion of Zip code availability. In checking out Kentucky, for instance, they found a gap of about 19 percent between the FCC's methodology and reality. The FCC reported 96 percent Zip code area availability of broadband; the GAO used data from a state alliance in Kentucky that showed 77 percent household availability. A Free Press report from August 2006 also notes that the FCC's so-called Zip code database is a private one that doesn't match US Census Zip Code Tabulation Areas, a standard method of defining regions. (The Free Press looks at a host of metrics, including the fact that Americans pay much more per Mbps for service, that that gap in payment has grown, and that the overall broadband speeds haven't increased comparable to other industrialized nations.)
Contrast this with Ofcom, the British spectrum regulator. Their 2006 communications market report is insanely detailed, and not uniformly congratulatory. There's some measure of non-political analysis present at all times. But the British have reason to cheer on broadband. Due to regulatory requirements that force BT to provide non-discriminatory access to its DSL lines, and to charge a uniform rate of about £11 per month for a naked line, Ofcom reports an actual number: 95 percent of homes passed by broadband DSL, 45 percent by cable. Further, broadband in this report is 512 Kbps. Still below the real threshold--1 Mbps symmetrical is much more reasonable when looking at what you can do and what other nations offer--but substantially higher.
Remember that both numbers are about availability, not about actual subscribers. Britain's 11.1m broadband subscribers represent about less than half the households. The 50m US broadband subscribers are somewhat further less than half. But let's recall: broadband in the UK must be at least 150% faster to be called that. More than a quarter of UK households use dial-up; less than a quarter in the US.
This has led a decrease of home telecom spending from nearly £120 per month in 2001 to about £80 in 2005 (in 2005 pounds). Broadband cost halved during that period. In one measure in the Ofcom report, 512 Kbps broadband dropped from £45 per month in 2001 to £10 per month in 2005--and "most operators no longer market a 512 Kbps service." 1 Mbps ADSL averages £15 per month now, and "greater than 1 Mbps" just £16 per month.
In the UK, it's now typical to be offered as fast as 8 Mbps ADSL at no cost beyond the £11 line fee--which BT may lower this year when they hit a milestone with regulators--if you subscribe to a certain mobile plan or a satellite television plan.
I can't find a similar time series in the US for the full package of services--the line, fixed-line calls, mobile calls, and broadband--but broadband prices have apparently stagnated.
I'm not proposing the regulation is the answer. But I'm pointing out the gap between what's commonly cited by reporters because the FCC puts the number out--99 percent of Americans and 95 percent of Zip codes--versus the reality, and how a comparable nation reports its numbers.
George Ou has some interesting points about 5 GHz, but there's more to the story: Ou hates the 2.4 GHz band with something greater than a passion--it's crowded, there aren't enough non-overlapping channels, and it's just so out of fashion. He's right about all that. The 5 GHz band has lots of possibilities, including 23 channels open for use. Before we get too excited, though, let me point out a few things that Ou didn't cover--primarily, the signal power output restriction for 5 GHz. 5 GHz 802.11 standards can't send signals as far as 2.4 GHz for a good hunk of the band. (23 channels are available for 802.11 specs; there are technically 24 possible in a different configuration.)
One of N's possible advantages of double-wide channels--instead of 22 MHz, they can use 40 MHz channels, which effectively doubles throughput. When you combine a newly efficient design for encoding, two or more radios, and double-wide channels, that's when you get the high symbol rate of 300 Mbps, with effective throughputs that could go well over 100 Mbps. The 100 Mbps throughput factors in--as I understand it--the expectation that N devices will have brief periods in which they can bond two channels.
Here's my executive summary so you don't have to read my entire analysis:
While 5 GHz is uncrowded and has more clear, non-overlapping channels that can be combined for the highest speeds in 802.11n, the rules governing 5 GHz for indoor use with omnidirectional antennas mean that only two double-wide channels in 5 GHz are available at anywhere near the maximum power (and thus potential range) of interior Wi-Fi. A number of 5 GHz channels offer power levels that are comparable to 2.4 GHz--and that assumes manufacturers will allow their 5 GHz radios to output enough juice to produce ranges similar to 2.4 GHz. But the lack of competing networks in 5 GHz could mean that 5 GHz N networks will almost certainly work much better in crowded RF environments--apartment buildings, for instance--than 2.4 GHz N networks which have very little chance to use double-wide channels. (Proviso: Some countries don't allow double-wide channels in 5 GHz.)
Ou says that "802.11b was supposed to have given way to its sibling standard 802.11a which operated in the 5 GHz range." That's not my recollection back at the time this was happening. 802.11b and a were seen as having somewhat different purposes: 802.11a had the potential for high-speed, short-range indoor connections, and long-range point-to-multipoint outdoor hookups. This was largely due to the higher cost of A, as Ou points out, but also due to the shorter range possible in indoor applications due to lower signal strength allowed. (Also, there were doubts about A's ability to be produced using CMOS chip processes, which Atheros put to rest, even as it moved into G for competitive reasons.)
He notes that while 802.11g boosted speed, "the problem was that in order to maintain backward compatibility, 802.11g also had to operate in the limited 2.4 GHz space and worse it had to switch to 802.11b if even one legacy 802.11b device joins the party." This isn't strictly correct as described, although it's widely stated this way. An 802.11b device forces a G network to drop down to B speeds only when the B device is transmitting or receiving data. There's a bit of extra network overhead, as well, if you choose for your G device to support B that does produce a single digit reduction in throughput even on G devices. The same will be true with N: older devices will occupy disproportionate amounts of time while transmitting or receiving, but if they aren't heavy network users, the N network should still work well.
I would argue that dual-band gateways didn't succeed in the consumer marketplace because of an absence of a compelling reason to use them. Throughput wasn't a big motivator. There were initially few adapters, none affordable, for 802.11a when 802.11g hit its stride. Remember Steve Jobs telling us 802.11a was dead? (Hey, their new device supports A, so I guess it got better.)
If you had mostly 802.11b/g devices and wanted to use A, you'd have to switch all your adapters over because no consumer devices supported simultaneous dual-band operation in 2.4 GHz and 5 GHz. Some enterprise hardware did (and still does) allow both bands either through two baseband chips or through two separate radios.
Ou complains that MIMO's leap into the market caused the lack of 5 GHz expansion. "Since it already required multiple radios for single band operation, adding an additional set of radio on the access point would have increased the already-high prices even higher." But there are few except very high end devices that would have duplicated radios; typically, as I just noted, it's one radio with two frequency ranges supported for either 2.4 GHz or 5 GHz operation.
Now here we get to the crux: Ou writes, "802.11n in my opinion should have NEVER permitted 2.4 GHz operation in the first place and should have only used the 5 GHz band." However, that idea has two flaws, one of which Ou admits and addresses. First, it breaks compatibility, which means that it's not a solution for people with B and G that want to gradually move to N. Ou suggests that a cheap single-radio could have been inserted into access points to handle B/G clients and the N system could have worked just in 5 GHz. Let's leave out all the radio engineering issues involved in that--like antenna coordination, cost of manufacture, firmware support, and so forth. The second flaw is more critical: 5 GHz has too many limitations in range.
The four spectrum hunks of legal 5 GHz channels each carry restrictions that don't dog 2.4 GHz, and that's why 5 GHz hasn't caught on. Where 802.11b/g/n devices can transmit as much as 1 watt of power at the antenna and use any channel indoors or outdoors, rules for 5 GHz prescribe in the U.S. and some other markets which channels may be used indoors only, and has much lower power levels for omnidirectional indoor use than 2.4 GHz allows.
There are now four bands in 5 GHz channelized for 802.11 in the US, although they're numbered somewhat strangely. In brief, there is total of 555 MHz across 23 channels in 802.11a/n. The lower four are indoor only; the higher 19 are indoor/outdoor. The lowest four (5.15 to 5.25 GHz) can have 50 mW of output power, the next four (5.25 GHz to 5.35 GHz), 250 mW; the next 11 (5.47 to 5.725 GHz), 250 mW; and the top four (5.725 to 5.825 GHz) up to 1 W. (There are further restrictions on 5.25 GHz to 5.725 GHz in terms of detecting and avoiding stepping on military radar transmissions, which share those bands. And the 802.11a spec specifies 40 mW/200 mW/800 mW instead of 50, 250, and 1,000, just to make it even more complicated.)
There are an enormous number of details about effective output, antenna gain, and so forth, but most of that affects the use of directional antennas and point-to-multipoint outdoor connections, not the use of interior omnidirectional antennas.
Because 5 GHz signals have shorter wavelengths than 2.4 GHz signals, at the same amount of power, they propagate shorter distances. They're also worse at penetrating solid objects. This means that even if you use the top four channels for 802.11a or single-wide channels for 802.11n in 5 GHz, you will only be able to send data less than half as far if that. There are only two double-wide channels possible in that top band.
In the 250 mW restricted range of 5 GHz, you could achieve the same range by using higher power in 5 GHz than in 2.4 GHz. But many of the devices that offer 2.4 GHz and 5 GHz radios don't compensate in 5 GHz by having higher-powered signal output. Thus a device that gives you 100 interior feet in any direction in 2.4 GHz could span less than 50 feet for this reason in 5 GHz. The lack of interference from competing networks could compensate for the shorter distance, however.
(Another issue: Some 802.11n device makers may not let you use double-wide channels in 2.4 GHz. Apple's new AirPort Extreme with 802.11n says in its advanced configuration manual--online already long before the product ships--that what it dubs the Use Wide Channels options is only available in 5 GHz. Conversely, Apple is promoting its AirPort Extreme with N in some European as only offering 20 MHz channels in 5 GHz because of regulatory limits. Thanks to Iljitsch van Beijnum for pointing to the manual and the European issue.)
Still, 5 GHz does offer some hope, and while Ou thinks the boat was missed, I see Apple's support in clients and adapters for 5 GHz in N, and Intel's support in its Centrino client for 5 GHz as a sign that that band will pick up steam. Note that Intel is certifying access points and routers with a Connect with Centrino label--and those devices will likely have to support 5 GHz, like this Intel-co-branded Buffalo router.
Let me reiterate one point here at the end: Manufacturers often limit their devices to 100 mW or even much less of output power for 2.4 GHz for reasons of cost. The problem in using 5 GHz will come entirely from whether those manufacturers decide to use the same power output limits for 2.4 GHz and 5 GHz even though they don't have to, or whether they'll actually take advantage of 5 GHz by boosting its power to put its range into parity.
On the heels of Apple's commitment to 802.11n, Intel gets in the act: The timing is no coincidence. Apple and Intel have been coordinating messages, and even though Apple won't be using Intel's chips--Atheros appears to be the anointed party--Intel wanted to wait until after the IEEE task group vote last week and after CES, too. Apple jumped the gun by a few days.
Notebook partners include Acer, Asus, Gateway, and Toshiba committed at this point, with systems available at the end of January using Centrino Duo. The Santa Rosa laptop chipset platform is due in the second quarter of 2007, at which point they expect more participation.
Like Apple, Intel says five times the throughput, twice the range, but only in comparison with their previous products--no actual specs on megabits per second or feet/meters. They do note that N will provide an hour more battery life under comparable circumstances with the previous Centrino flavor. They will support 2.4 GHz and 5 GHz N channels.
The announcement includes the Connect with Centrino program, in which access point makers commit to rigorous testing with Intel to ensure interoperability. This is clever, because it will go beyond the Wi-Fi Alliance tests, which focus on wireless protocols, and include the whole ecosystem of DHCP addressing and other factors. Asus, Belkin, Buffalo, D-Link, and NetGear are all part of this first wave of branding.
I'm not trying to be incendiary, but that's now the projected ratification date: Ephraim Schwartz files this detailed column for InfoWorld about the steps between now and then for moving 802.11n out of Task Group N and into the full body. Draft 1.10 was approved the other day; with minor changes that goes out as 2.0 to a larger voting pool. The odds are about zero at this point for any changes that will involve anything but firmware upgrades as there will be an entire entrenched industry for Draft 2.0-based products.
The final ratification isn't expected until October 2008, but the spec will be nailed down by around January 2008, which was about the original plan as of a few months ago.
Recall that 802.11n was expected to be ratified in 2006 a couple of years ago. The roadblocks that prevented that have now been overcome.
A San Francisco Chronicle editorial asks the Board of Supervisors to show an alternate Wi-Fi plan: The editors think that with EarthLink on the hook, minor quibbles should be swept away unless the board has a viable alternative plan that wouldn't involve an "experiment" with $10m of taxpayer dollars. It's a tricky position. San Francisco has received about as good a deal as any other city its size. Portland, Ore., is the sole exception, with free 1 Mbps ad-supported service from MetroFi. SF rejected MetroFi's proposal, and it would be good for the larger Wi-Fi community to see different models built in different large metropolitan areas to see which one works best, or if they all work equally well or poorly.
MetroFi enters its second phase of coverage of the City of Roses: Portland, Oregon's free-with-ads Wi-Fi network expands into Old Town, the Pearl District, Portland State University's area, and Southeast Portland over the next four months. The Pearl District is better known as "the part of town that Powell's Books built." The initial rollout covered about 5 percent of the city. MetroFi said that 17,000 hours were spent online during December by 3,000 uniquely registered users. Registration is required for network use. The network should be complete by mid-2008.
A related phase of MetroFi's revenue model has just started, too, The Oregonian reports. A city office in Northwest Portland will begin paying "less than $200 a month" for a wireless link. Selling high-speed dedicated links to city offices and businesses is a key part of EarthLink's model, but I haven't heard MetroFi mention it much.
An article in the Portland Tribune seems to take MetroFi to task for its current coverage area and the cost of the bridge generally needed to connect indoors--the paper says that the city was saying $50 to $80, but MetroFi's recommended bridge is $120. It's pretty anecdotal. That doesn't mean it's not true, but I'd be curious to see if someone performs an extensive wardrive in the covered areas. You can always find people who won't get coverage, especially with such a small area deployed; it's much harder to find people who are perfectly satisfied.
Also, Michael Weinberg, the fellow quoted in the lead of the Tribune article, is heavily involved in Personal Telco, the veteran community wireless group in that city. While not in competition with MetroFi, Personal Telco had argued for a different plan for the citywide network--not one that they would profit from, let me note, but one that focused more on community than a private operator. That fact should have been noted by the reporter. His points, however, are still well taken. Paying $120 for a "free" service, even as a one-time fee, has apparently not been well explained. This has bitten other early metro-scale Wi-Fi projects, too, where "free" or "ubiquitous" weren't necessarily coupled with "paid bridge."
Adam Boettiger, a colleague of mine from long ago, is pictured in the article's opening. He's got an Apple laptop there, and if it's a Core 2 Duo, I'll be curious whether he gets better indoor reception when the 802.11n enabler from Apple ships in February. The enabler will allow many existing Macs to suddenly have N features, which should add better receive sensitivity and transmit power when both radios in the 802.11n chips are turned on.
Big news from the IEEE: Draft 2.0 of 802.11n moves forward: The vote was 100-0 with 5 abstentions, Matthew Gast notes from the London meeting. This is a significant milestone from a lot of different directions. It's one thing to achieve a 75-percent supermajority necessary to advance a draft into its final stages, which is just the tweaking and fixing that brings it to ratification with no significant alterations. It's another to get the bag of cats that are the stakeholders in the task group to vote unanimously.
Draft 2.0 now goes out for letter ballot, where IEEE members choose to move the process along further, but the vote to take the draft into that status indicates there should be little problem with the formal adoption as the final basis of 802.11n. (Technically, Draft 1.10 was approved, and if the letter ballot succeeds, the new draft will be numbered 2.0.)
This also means that it's extremely likely that my concern over the last nine months about early Draft N equipment not all being upgradable through firmware to a Draft 2.0 and final release standard will prove misplaced. With the industry providing no hardware upgrade guarantees--like, "We'll swap your gateway if we can't make it as fast as the standard says and fully interoperable"--I've been dubious about early adoption of Draft N, especially that equipment based on Draft 1.0. Further, most early gear has revealed remarkable inconsistency in performance and interoperability with even like devices in testing by magazines and online publications.
The vote today indicates that there's a very strong direction for 802.11n, and I expect within a matter of weeks that we'll see waves of firmware upgrades for existing products, real availability of Draft 2.0 chipsets--Airgo wasn't the only one working in anticipation of this accepted draft--and a timetable for the Wi-Fi Alliance to certify Draft 2.0-compliant devices.
After four years, this is a big step forward.
Update: Atheros released a statement from their CTO that Atheros's Xspan chips can be made compliant with today's approved draft via a software update.
Broadcom's Bill Bunch also said in an interview, "All of our products that we've ever shipped since Day One will be upgradable to the final draft 2.0." It's all about firmware upgrades now.
TG Daily says Intel will announce Draft N chipsets next Tuesday: The site writes that the Kedron wireless chipset may already be in finished devices, and Intel's announcement could mean they have equipment from partners to show with the Draft N standard enabled. Kedron will work in conjunction with Santa Rosa, the next-generation laptop chipset, which is due in second quarter 2006.
Just off the phone with Apple, which says Mac owners with 802.11n-capable hardware will be charged a $1.99 fee: A spokesperson with Apple provided me with a response that explains that the updater will be available for purchase from their online store at a "nominal fee" in order "to comply with Generally Accepted Accounting Principles for revenue recognition, which generally require that we charge for significant feature enhancements, such as 802.11n, when added to previously purchased products."
The company will eventually fully enable 802.11n on all Macs with appropriate hardware that they ship. The spokesperson couldn't yet provide a timetable for when that switchover would happen. If you purchase a Mac with Draft N chips between now and then, it's possible you would need to pay the $1.99, but, frankly, you'll just find someone with the enabler, won't you?
The enabler will be included on the new AirPort Extreme Base Station's CD.
The concept of municipally owned Wi-Fi makes a comeback: I've been blunt in noting for a couple of years that I don't generally support the idea of cities entering the telecommunications business except in cases in which alternatives are few. But municipally owned wireless really doesn't mean that towns and metropolises are encouraged to develop, say, a utility-based in-house expertise. Rather, it's the other side of the line from public-private partnerships. And it requires acceptance of the notion that substantially more "free" Wi-Fi access is a public good that's more valuable than outsourcing the network and allowing profits to accrue to private enterprise.
The Institute for Local Self-Reliance issued a report that argues for a particular form of publicly owned network in which the entity builds infrastructure which it resells on non-discriminatory, broad, wholesale terms. This is distinct from many city-proposed and city-built networks that are either entirely free to use or which put the city in the role of the main or sole retail purveyor of service. Tacoma Power's broadband network has been on this basis since at least 2000, with multiple residential and business providers selling access and the Washington utility not involved in that at all. (Tacoma Power sells cable television service directly, however.)
San Francisco, meanwhile, saw a report issued by the San Francisco budget analyst's office for the Board of Supervisors, a group that has been generally critical of the Wi-Fi plan championed by Mayor Gavin Newsom, and that will shortly reach the supervisors for a vote. The analyst found that a city-owned network could produce a deficit or profit, depending on a variety of assumptions, all of which had the city using such a network to upgrade or replace existing telecom and data contracts--which is not assured in San Francisco's contract with EarthLink. San Francisco has 642 T-1 lines, for instance, and replacing those with in-house wireless links (where fiber wouldn't be available) would save $5,346 per year beyond the setup cost. However, the analyst's report is highly critical of the process by which EarthLink was selected and, by extension, Google incorporated as the free, lower-speed provider. One specific complaint is that no feasibility study was conducted before issuing the RFI and then RFP and then awarding the contract. (Report in PDF form)
The largest hotspot infrastructure operator says they've had 30m paid connections ever, 14m in 2006: In a press release that's not yet online, Wayport trumpets some numbers. They now operate 13,000 hotspots worldwide, with the majority (my words, not theirs) comprising McDonald's locations (direct contract), and locations they manage for AT&T FreedomLink, such as at the UPS Store.
The math is interesting, though, as 14m divided into 13,000 locations means just over 1,000 connections per location or three per day on average. Now some locations will be very high traffic, which means that many hotspots might see no more than one connection per day. And the settlement rate for roaming runs from 50 cents to $2 per connection, from what I have been told over the years. So 1,000 connections, even with a healthy proportion of walk-up $7 to $12 per day rates, might mean only $2,000 to $4,000 in revenue.
But that's me being pedantic. Why? Because it's irrelevant for most of their locations how many connections are made. Really? Let me explain.
McDonald's, which represents 8,000 of the 13,000, uses the Wi-Fi and Internet network for its own purposes, which they are apparently happy with to judge by public comments and the deal's ongoing nature. Incremental business is great, but a few additional daily users is icing on the connectivity cake.
A few thousand other locations are operated on behalf of AT&T FreedomLink, a firm that uses its network as an additional incentive to its DSL customers to remain DSL customers. For $2 per month, AT&T DSL customers gain access to McDonald's locations and AT&T FreedomLink home network hotspots. (Wayport's hotels and other locations add $20 per month for use.) This, again, means that daily sessions and actual revenue is relatively uninteresting as opposed to customer loyalty and a long-term strategy. Further, AT&T has several airport locations which would produce session usage and walk-up fees orders of magnitude higher than a UPS Store outlet.
This leaves Wayport's 1,000-odd hotels and other properties, in some of which, Wayport is paid by the hotel (such as Wyndham) to provide high-speed Internet access at no cost to its loyalty-program guests.
So those 14m sessions in 2006? Less important on a per-location basis and more important as a measure of overall usage across the networks that Wayport runs.
Reviving upstate New York includes broadband boost: Outside of metro NYC, there's a paucity of broadband availability and relatively low uptake. Governor Spitzer wants to change that, following plans that already bore some fruit during the Pataki administration. The state will try to use its physical resources to help gives incentives to broadband providers. Spitzer wants "near universal" coverage by 2010. A related article rounds up some wireless options on the way in the Ithaca area. [links via Muniwireless.com]
Long Island issues RFP for network: The counties of Suffolk and Nassau are looking for responses by March 19, 2007. Muniwireless.com notes that 2.7m residents would be passed across a 750-square-mile area. Requirements include universal wireless broadband access. The counties hope to boost economic growth, improve government services, add to educational efforts, make Long Island better for tourists and business travelers, increase competition for broadband, and be ready to extend access to transit. (Plan details here, although the leading statement that Wi-Fi stands for wireless fidelity is like saying Pink stands for insulation.)
Rhode Island issues plans for statewide rollout: The Business Innovation Factory that's driving the state-wide network has set its plans after eight months of testing. They expect the network to cost $28m for buildout and thre years of operations, at which point their projections show the network to be self-sustaining. The group is looking for the state to back guarantees to use private debt to fund the project. It will take 12 to 18 months to build out.
Steve Jobs directly confirmed to one questioner that Apple would charge for its 802.11n enabler for existing Macintoshes: A reader who prefers to remain anonymous forwarded me the mail he sent to Steve Jobs, Apple's CEO, and the reply he received. He included mail headers so that I can confirm the mail is legitimate. The reader asked Jobs whether press reports were in error that Apple would charge $5 for an "enabler" that would turn on the 802.11n functions in most Core 2 Duo and Xeon systems shipped in 2006. (The 17-inch 1.83 GHz iMac with Core 2 Duo lacks the necessary chip.)
Jobs replied, simply, "It's the law," which would confirm that the Sarbanes-Oxley requirement that seemed bizarre to me is, in fact, correct. In several reports, the law is cited as requiring different accounting for earnings on products that are shipped and later provide new functionality that wasn't initially advertised. Charging for the updater means that the functionality didn't come for free. I still hope to hear some better analysis about why the law requires this kind of product update micromanagement.
In any case, the email is legitimate, and Jobs's reply is unambiguous.
My earlier post on this was titled, "Apple Won't Charge $5 for 802.11n," but what I said--not so clearly--was that Apple would only tacitly charge that if they charged anything. The company isn't discussing releasing a locked, serialized enabler that works only with laptops and desktops that have been approved for update. Rather, they may charge $5 for an enabler, but the enabler will quickly be distributed for free, however informally, until the millions of older machines are patched.
I suspect based on Jobs's response, too, that Mac OS X 10.5 (Leopard), which will be a paid operating system update shipping in second quarter 2007, could include the enabler, too, since that's a separate fee. The AirPort Extreme Base Station with 802.11n will include the enabler, and I'm extending the logic from there.
CES reveals that home media networks will require wired, wireless components: I know this might seem like an obvious conclusion, but the buzz around 802.11n, ultrawideband, and even ZigBee makes it seem like manufacturers think that wires are just too, too passé these days. Not the case, writes David Haskin in Computerworld. CES revealed that the makers of the gear that will link computers and home entertainment systems--not to mention the PC and CE makers themselves--see room for radio waves and copper. For instance, Sharp's 108-inch LCD TV has HomePlug AV support--a 200 Mbps electrical wiring standard that might finally come into its own this year--and no wireless. And network-attached storage for media will likely take a wired path with wireless options.
The San Francisco Board of Supervisors could vote down contract: The supervisors are a notoriously independent bunch, and have been critical for months of the way in which the mayor has proceeded down the metro-scale Wi-Fi path with EarthLink (directly) and Google (only slightly less directly). EarthLink's muni head Don Berryman is reported to have handicapped the supervisors vote as four to five in favor, three against, and three to four undecided. Six votes are needed to approve the contract that was finalized several days ago. A vote may happen as early as February.
Berryman made his comments in a public meeting as EarthLink lobbies for support from the community. The Red Herring reports that Berryman said that the company had no votes against it in the six cities in which it has started deployment, which is a remarkable record. The company has 17 municipal deals settled or in negotiations, and has built 250 square miles of network to date. That number could 5 to 10 times higher by the end of 2008.
I'm pleased to see that Berryman acknowledge that two of the most widely predicted factors outside of the industry building Wi-Fi networks turned out to be stumbling blocks: real-estate rights and access point density. The former is an argument that folks promoting other network technologies have made repeatedly: It's not cheap or easy to obtain the rights to thousands of poles and rooftops, even if a city owns the utility and is backing you in your efforts.
On the latter topic, recall that the number was down in the low 20 APs per square mile estimated to build municipal service in 2005; EarthLink most recently put the number at 35. The more APs, however, the lower power each AP operates at (to avoid significant overlap), which in turn reduces the noise floor for "competing" devices.
Dianah Neff, a consultant with Civitium and former CIO of Philadelphia--Civitium put together San Francisco's bid proposals--said that 3,000 300 municipalities are now examining building networks in the U.S. Update: Neff says she was misquoted as stating 3,000; she said 300; see comments.
Washington Post's Rob Pegoraro, IDG's Nancy Gohring are sucked in by cordless power: Pegoraro describes spotting two wireless power firms that he describes as more plugless than wireless. Powercast can "broadcast" power over three feet to circuits that could be integrated into existing electronics. Fulton's eCoupled, by contrast, uses charging surface: an equipped device, when it touches an otherwise safe charging surface, trickles in power.
My good friend Nancy, meanwhile, arrives at the end of 17-hour trip from Ireland to Las Vegas, and finds one missing charging cable and one broken one. She offers more detail about eCoupled. The company expects travelers could bring a small pad with them that plugs into AC power and then place their various devices on top, avoiding the need to bring cables of any kind. Motorola, Visteon, and Herman Miller are all working with eCoupled, though no products are announced.
The chipmaker CSR wants to preserve existing Bluetooth business by offering GPS as a cheap, incremental improvement: CSR says that it will cost about $1 to add a GPS receiver in a combined Bluetooth/GPS chip, and that the chip--make possible by its acquisition of two GPS firms--will have far higher sensitivity than other chips on the market. E911 service in the US requires some kind of automated location service be embedded in phones. As location services are now being sold by carriers based on their cheap, embedded GPS receivers, CSR may have a market in providing better positioning, a lower bill-of-goods, and better battery life.
JiWire facilitates free hotspot access via commercials that must be viewed in their entirety: The idea is that the Ultramercial, the brainchild of the eponymous company, has a high enough value to advertisers because a hotspot visitor is forced to view the ad before being given access. While JiWire, a company I own a very tiny piece of, isn't disclosing the ad rates it obtains from companies using this commercial-for-free-service format, they did note a 7-percent clickthrough rate versus industry averages of 1 percent for all online advertising.
I can't provide more information than is available publicly, but I can note that it's relatively easy to look at what a hotspot operator might obtain through a roaming agreement or as their net cost for pay-as-you-go and consider whether this format might deliver more users and thus incremental revenue that make advertising the right mix.
This form of advertising is somewhat different than the 1-inch advertising bar used by MetroFi to fund its cost-free service in that the expectation with an Ultramercial is that someone is giving up a small piece of their time in exchange for a high-value item. The ad bar used by MetroFi, as one example, is background advertising that's continuous, and thus has a much lower value relative to its exposure.
More bad news for customers outside of cities: Verizon will lose 1.6m phone lines, 234,000 data customers, and 600,000 long-distance customers in Maine, Vermont, and New Hampshire, and own 60 percent in the new venture formed with FairPoint. FairPoint operates just 308,000 phone lines today, and is generally the incumbent local exchange carrier in these mostly tiny exchanges of less than 2,500 phone lines. By spinning these lines off, Verizon gets to take some of the cost of operating rural exchanges off its books, and the new entity may benefit from universal service fund fees, which provide significant revenue to rural exchanges.
Sprint Nextel spun off its access lines into a company called Embarq last year, as did smaller telco Alltel, which sold its lines to Valor Communications, which operates the new entity as Windstream. FairPoint, Embarq, Windstream, and a fourth firm, CenturyTel (a previous buyer of rural Verizon lines) represent a large, consolidated hunk of rural callers.
These markets may be ripe for more wireless, as wireline services have enormously lagged in these areas compared with the rest of the US. (Remember that the FCC counts broadband as 200 Kbps in either direction, and it counts a Zip code as having broadband service if there's a single provisioned line and a single customer for that line.
Sprint Nextel and Clearwire have licenses that likely cover a large rural area, and the question will be whether they can find it cost effective and efficacious to deploy for those underserved residents.
We all know that European Wi-Fi access is overpriced: Articles appear practically every week from some commentator--either a traveler into Europe or a bona fide European--rails against hotel and hotspot Internet pricing. Free-Hotspot.com thus has scored a minor coup in building a network of over 600 locations in 14 countries that are free, free, free. They say 40,000 people use the network each month. The group supports free service via advertising. The company says that free hotspots overall doubled in Europe in 2006, but given that they claim 60% of that growth, that means there are no more than 1,000 locations that they're aware of that don't charge a fee for access. In the US, that number is as high as 10,000 across a similar population.
Breathlessly reported worldwide, Apple has been said to be planning to charge $5 for the 802.11n updater for their Core 2 Duo Intel Macs. Pshaw: Everyone's looking for previously unreported news angle on Apple, and a story that seems to have started at iLounge on Jan. 10 seemed to spread across the globe in a day. The item is that Apple will charge $5 for an updater for 802.11n on machines that can be updated. The claim from unnamed "Apple representatives" was that the Sarbanes-Oxley Act will require a fee to be charged to enable features in a product that's already shipped to avoid restating earnings on that product.
Which seems insane to me as the logic for the fee. But as the newsgroup posters say, IANAL (I am not a lawyer), and thus perhaps there's an SEC memo on this that I don't know about. (Post in comments if you have more knowledge.)
All but one Intel Mac model shipped in 2006 that contain Core 2 Duo or Xeon processors can be upgraded using what Apple dubs an "enabler"--a firmware patch coupled with updated AirPort software for Mac OS X--that will be on the CD that comes with the new AirPort Extreme Base Station with 802.11n that's scheduled to ship in February. (Some reporting incorrectly states that Apple said only Core 2 Duo models can be upgraded.)
I asked Apple if the $5 fee was correct. In some limited cases, Apple has charged a fee, often as high as $20, to ship new operating systems discs or an updater that was otherwise available for download. An Apple spokesperson declined to comment on "rumors or speculation," but reiterated that the enabler would ship with new AirPort Extreme Base Stations next month.
Without reading between the lines, it's pretty clear that Apple implicitly told me the enabler won't be restricted. Why? Because by including the enabler on a CD and not stating any limits on upgrades--a system admin could upgrade 100s of Intel Macs with a single enabler--there's way to prevent that enabler from instantly being spread via BitTorrent, email, and so on. They'd have to install some kind of serial-number-based registration system, a tracking methodology via the Internet, and other customer support.
If Apple is interpreting Sarbanes as requiring a fee, they're also going to make it very simple to avoid that fee.
More likely, I suspect that Apple is readying a Mac OS X 10.4.9 update, the final minor release for Tiger, and that the timing of the release of the 802.11n enabler and the updated AirPort Extreme will be tied to 10.4.9. It would make sense that 10.4.9, incorporating newer AirPort software for managing the connection, would be required to use 802.11n.
Alereon and Stonestreet announce integrated ultrawideband (UWB) for Compact Flash, Secure Digital: The combination of software and hardware can be used with Windows Mobile and CE. This would allow makers of handheld devices and smartphones using the operating systems to add UWB without redesigning their products. UWB had a kind of coming-out party at the Consumer Electronics Show last week, with devices shown that had never before seen the light of day. Actual shipping UWB products are expected as soon as this month, at long last.
JiWire quietly released a Skype agent that can answer hotspot queries: This Skype agent has the automated intelligence to answer questions about hotspot locations using JiWire's ginormous worldwide directory. "Airports in Malaysia?" I ask. It replies, "I found 8 airports with wifi near Kuala Lumpur," and lists them. If I enter my email address, the results are sent. (Disclosure: I have a very small interest in JiWire.)
Task Group N in the 802.11 Working Group is expected to approve a draft for balloting this week: The IEEE is a consensus-driven standards group, so its process grinds slowly. This week, approval is expected for the draft that will ultimately become the final 802.11n standard. The draft needs first to be voted on to proceed to a letter ballot, which is collected and reported for the following meeting in March. I'll have news when it's reported.
Wired reports on the gas pump with extra octane--I mean, octal: The Ovation iX has a touchscreen, speakers, and Wi-Fi built in. It shows commercials, but can also transfer music files to equipment in a car that's appropriately configured. Microsoft's Automotive Business Unit was involved in the development. Of course, I totally want to be waiting for gas while somebody fiddles with settings to buy music from a pump. Pump rage could be 2008's road rage.
Powerline devices don't work with each other: This is preventing the kind of growth predicted a few years ago as newer devices that use home electrical wiring as a network media were to hit the market. With specs like that of the HomePlug Alliance offering speeds of 200 Mbps, you'd have expected something to coalesce around a dominant flavor. Instead, you have chipmakers each pushing their own solution. This year's CES doesn't seem to offer clarity. Nancy Gohring reports for IDG News Service that the dominant retail manufacturer NetGear is actually unhappy with the lack of interop because it's stifled the market as a whole of which they'd like to have a piece.
Wi-Fi Planet offers a short but thorough look into the concerns about Wi-Fi radios' effects on human health: The article includes comments from several researchers, all of which agree that the signals from wireless LANs fall well below any minimum level of concern based on decades of research, and that specific projects used to measure electromagnetic radiation confirm Wi-Fi's small role in an overall hum of background noise.
Strip away the high-pitch, touch-screen display, and what makes an iPhone unique? I spent the week at Macworld Expo in San Francisco talking to colleagues and others about the iPhone, announced by CEO Steve Jobs on Tuesday morning, and not shipping until June. The consensus is that while it is one of the coolest pieces of technology on the planet--even though you can't get one--it's not that tremendous of a phone as such.
It's more that it's a great packaging of several technologies in what appears to be a seamless, consistent platform. It should be easier to use than any existing smartphone or cell phone, handle music and video and photo display infinitely better than any such phone, and be a better Internet-connected device for browsing, and perhaps for email. (It has a rich HTML email client.)
But let's face it: the phone features are rather dull, with the exception of so-called random-access voicemail, when compared to the music and Internet access features. You can use an attractive address book to place calls, your SMS messages are organized into an iChat-like display (without actually being an instant messaging client), and you can send email and carry out other tasks while you're making phone calls. Random access voicemail lets you see a visual display of messages by caller (if they provided Caller ID and are in your address book, ostensibly), and then click to listen to any message without having to skip through others. That's rather cool, and it's more efficient than today's system by a few orders of magnitude.
Where I think the iPhone has the potential to change the face of telephony isn't on the side of usability. It may. If successful, it may force interface and feature redesigns by platform developers and handset makers. But it's big impact could be on converged calling, in which Wi-Fi and cell networks are used interchangeably, to the benefit of consumers and cell operators. This is commonly called unlicensed mobile access (UMA), a name that describes a particular technology rather than the concept, but UMA is the only such approach currently deployed. T-Mobile offers it in Washington State in the U.S.; BT just rolled it out over Wi-Fi--they previously had a Bluetooth-only offering--to its broadband subscribers in the UK.
A couple of colleagues reading my earlier post on the iPhone wanted me to clarify this point. The iPhone, as described by Jobs, will seamlessly roam data connections among EDGE and Wi-Fi networks. But it cannot make phone calls over Wi-Fi, and it's a closed platform that may require a process in which software gets approved for installation--meaning that Skype and Vonage could be out of luck. (The other five smartphone platforms, counting many different versions of Linux as one platform, allow developers to write and offer software for installation, although not all software can access all phone and cell network features.)
The iPhone would be Cingular's first smartphone that incorporates Wi-Fi in a deep and meaningful way. There are a number of smartphones in the US and worldwide that offer Wi-Fi, but typically only for certain subsets of tasks and only UMA offerings allow calls over Wi-Fi without the involvement of third party software. And those third-party programs can't tie into the cell network.
Back when AT&T was SBC, they were already talking about Cingular offering some kind of Wi-Fi calling. One statement in 2004 pegged a rollout for early 2006. Didn't happen, of course. But now the combined AT&T owns 100 percent of what will now be called the wireless division of AT&T--the Cingular name starts to die Monday. And the combined AT&T also has tens of millions of DSL customers across its territory. They also have thousands of Wi-Fi hotspots which they resell for cheap access to their DSL subscribers ($2 per month). This gives them the perfect audience to sell converged calling to using the iPhone as part of the bait. "Get the coolest phone in the world, and unlimited calls over Wi-Fi along with television and broadband!" Something like that. Offloading cell minutes to Wi-Fi cuts the cost enormously of fulfilling voice service.
The reason that DSL plus converged calling makes sense is that the fixed-line operator can provision VoIP on both sides of the DSL connection. On the home side, they sell you a Wi-Fi router with voice prioritization (Quality of Service via the Wi-Fi Multimedia or WMM extension), and have you put that router as the first device on your network, plugged into your DSL modem. (This is how T-Mobile offers UMA today in Washington State--they suggest you buy the router for best Wi-Fi and network performance; their router also preserves battery life through WMM Power Save.)
On the central office side, where the DSL line from your house terminates, the broadband operator can transport voice packets off to the voice network rather than transit them over the public Internet. A provider like BT or AT&T--or even a competitive DSL carrier like Speakeasy Networks, which I use for VoIP at home and work--can offer something very close to a circuit-switched voice line, rather than something like Skype or Vonage. It's not Internet telephony, it's provisioned VoIP.
The iPhone is just a portent, and Cingular could just as easily offer the same Nokia, Samsung, and Motorola GSM-based UMA phones that BT and T-Mobile have. But I was frustrated when I tested the Nokia 6136 with T-Mobile's UMA plan. It has Wi-Fi, but it has no browser (not even a crummy one), and makes no real use of Wi-Fi. The iPhone can richly employ Wi-Fi and cell data networks by having essentially a desktop operating system with desktop-class applications. Combine that with seamless, best-available calling, and unlimited calling over Wi-Fi, whether at home, office, or at a hotspot and you've got a plan that could attract those interested in far more than the cool factor.
The network we've all been waiting for: EarthLink this week finally fulfilled the first phase in their buildout of Philadelphia, the largest urban deployment currently underway in the the U.S. While other networks are larger in scale, and larger urban networks are planned, Phila. is the one to watch because of its prominent political role in the development of arguments for and against municipally authorized or built Wi-Fi networks.
EarthLink committed from the beginning of its negotiations with Phila. to a test network, and it's been since about May that the company started down the path of this first stage, a 15-square-mile network in which they will prove their approach works. Once a review of this pilot has been achieved, the company will proceed to build the first network.
Use of the network is free until Jan. 21.
Kodak offers 8-inch, 10-inch Wi-Fi enabled digital picture frames: The "picture frames" can show JPEGs and a several movie formats. The frames will ship in March for $280 and $220, respectively. An 8-inch version with USB and a card reader but no Wi-Fi is $180, and a 7-inch unconnected version $130. Several frame plates are available for variety's sake, including a wipe-off whiteboard version.
Apple will ship AirPort Extreme with Draft N in February: The former-computer company--renamed from Apple Computer to just Apple today--has gone Draft N happy. Their version will have no new, special name, and support both 2.4 GHz (802.11b/g/n) and 5 GHz (802.11a/n). Computers shipped to date with Intel Core 2 Duo chips (with the exception of a single iMac model) can be upgraded to add Draft N capability through a firmware patch due in February along with the updated base station. The new AirPort Extreme will cost $179. It looks like Apple will not refresh its Extreme Card as all new Macs and other Apple gear, like today's iPhone due in June, will include 802.11n built in. The device has a new form factor, with similar width and height to a Mac mini and Apple TV.
The iPhone will support EDGE and Wi-Fi: The iPhone is months from shipping, and will run $499 or $599 (4 GB or 8 GB) at introduction. The subscription plan cost from Cingular Wireless, its exclusive reseller, wasn't mentioned. The iPhone will automatically switch between Wi-Fi and EDGE for retrieving data as networks are available. With a fully enabled Web browser built in, iPhone users will be able to access hotspots that other phone users can only reach with an effort or not at all. There was no mention of converged calling over Wi-Fi and cell, or even VoIP applications--which makes sense given Cingular's voice focus. Apple said you cannot make calls (at this point) over Wi-Fi. That may change by introduction, or may be a future plan as AT&T plots Cingular's converged future.
Finally, the Apple TV, the release name for the codenamed iTV, will stream and store content: The $299 device, shipping in February, has Ethernet, 802.11n, HDMI (high-def audio/video), and composite video, as well as analog and optical digital audio outputs. It can synchronize content with one networked computer, Mac or Windows, and stream content from five networked machines. It sports a 40 GB hard drive.
I'm here at Macworld Expo, and will file more Wi-Fi-related news if it comes up.
Vonage signs three contract to resell EarthLink's metro-scale Wi-Fi networks its voice customers: Interesting synergy, because it's a wholesaler-retailer relationship, not exactly a strategic partnership, because EarthLink is generally constrained to offer non-disriminatory wholesale access to all its metro-scale networks by contract with the cities it is working with. While EarthLink has its own VoIP service it's been selling, and is looking to expand, Vonage is a great reseller of this kind of service, as they will be able to leverage offer Wi-Fi-based VoIP phones this year with the greater coverage of a citywide network.
This highlights, once again, that people will need per-human-being or per-family accounts that many devices can access at different rates than full-price-per-device. This is Devicescape's whole point of its new service, and its only becoming more obvious over time.
Wi-Fi Protected Setup (WPS) will use push-button or PIN to simplify secure network setup: The Wi-Fi Alliance has rolled out its long-expected and openly discussed WPS system for its members to choose to implement and have certified as a component of their Wi-Fi devices. With WPS, a router can provide strong WPA/WPA2 encryption keys to client devices with the push of a button or the entry of a 4- or 8-digit personal identification number (PIN). The PIN can be generated by software or preprogrammed into a client device and printed on an included card. Push-button authentication will work essentially like Buffalo's AOSS method; PIN authentication like Atheros's JumpStart (which Atheros notes is compatible in version 2.0 with WPS).
With WPS, however, the system will automatically set the SSID or network name as part of the setup process. That's both for simplicity's sake, and for security reasons. WPA/WPA2 uses the SSID as part of the process by which it generates the encryption elements used to protect network traffic. There's been some speculation that default networks name like "default" could be paired with massive, precalculated databases of short dictionary-based passphrases.
Products with WPS enabled should be available in the first quarter of 2007, but the Alliance announced several pieces of equipment--mostly reference design from chipmakers--that were certified as part of the process of finalizing the standard. This includes designs from Atheros, Broadcom, Buffalo, Conexant, Intel, Marvell, and Ralink.
Future releases will support near-field communications (NFC), in which one device is held near another to initiate a key exchange; and USB-based support in which a memory stick is inserted into a gateway or computer, and then inserted in each client device to add them to the network. Wi-Fi Alliance managing director Frank Hanzlik said in an interview that there was backing for all four forms of WPS association because "you might have vastly different products over time that might want to be able to capitalize on this." Some equipment may have no display whatsoever, like putative wireless headphones that might have a volume adjustment and nothing else.
WPS should eliminate the awkward entry of WPA passphrases now required on devices that lack keyboards. If I never have to use a directional arrow or a phone interface to "type" letters again for a passphrase, it will be too soon.
Hanzlik said that WPS provides a standards-based method for its members to implement simplified security, although each member may choose a different user interface to wrap around the process. He noted that WPS would tend to remove security as a differentiator among products, but that vendors appeared ready to do so, based on the broad participation in the Alliance's development.
This is the first standard to come out of the Alliance, rather than the IEEE or other groups, Hanzlik noted, because the particular combination of factors--which include user experience and usability--weren't under discussion in other industry forums. The standard will be available for download for a nominal fee. WPS is also unique in requiring WPA or WPA2 support in all devices. "By design, this does not support WEP," Hanzlik said.
WPS's timing is good for equipment makers because of the near-term gelling of the IEEE 802.11 Task Group's Draft 2.0. The Alliance will have a certification program for Draft 2.0. Hanzlik noted, "That is exactly what we hoped--that we'd have this out there with enough time that all the folks getting ready to bring out the new 802.11n draft certified products come June would be able to incorporate this."
Hanzlik expects that manufacturers will offer WPS upgrades for some older client equipment, as PIN authentication for WPS could be handled through existing client software and Web-based gateway configuration. The Alliance would likely not separately test legacy equipment that receives upgrades, but all new devices that want to sport WPS would go through certification.
The 48-page agreement is actually rather vague in many places: Most surprising to me--perhaps I've been missing this detail--is EarthLink's agreement to pay a five-percent franchise fee to the city for access to rights of way (ROW). These fees have rarely appeared in other agreements, in which yearly pole fees or other arrangements have provided a different method of taxation. EarthLink will pay $600,000 in advance against ROW fees.
Google is not mentioned by name in this agreement. They are the Basic Service plan provider unless they've disappeared from the deal, a fact I have not heard mentioned yet.
EarthLink has to build a test network up to two square miles in area, and then achieve approval of characteristics before proceeding. The contract mentions a vague notion that EarthLink will work with the city to deal with any issues of interference. But there are no actual steps for alleviation that are mentioned. EarthLink has to keep its network up to date with comparable cities in terms of equipment used. The agreement lasts four years starting at a certain point in deployment and can be renewed with mutually agreed-on changes for two additional four-year periods.
The agreement has a 180-day clock from execution for the Board of Supervisors and the Public Utility Commission to approve the contract and the pole agreements, respectively. After 180 days, EarthLink can choose to walk away.
EarthLink must sell wholesale access on a non-discriminatory basis. Roaming is required, but individual ISPs have to sort out how they allow users from the SF network to roam onto other networks. There is no mention of whether deposits can be required from potential retail ISPs using the network.
While it's unclear whether the privacy provisions will assuage those who were concerned about Google and EarthLink's plans for user information on the network, the terms of information use do seem reasonably protective. Personal information (as defined in the document) can't be shared without an opt-in process to anyone but EarthLink suppliers and marketing partners, from which communication users must be allowed to opt out. Location data will be collected, but you can opt out, and it will be retained only 60 days. Law-enforcement needs and regulations will trump this, natch.
The agreement defines several levels of service. Basic Service is 300 Kbps symmetrical, but noted as "best effort." That will be free. Some minimal user information is needed to use that level of service. The city can ask for changes to the service, but EarthLink doesn't have to make them. The service speed must increase yearly to 15% of the "best selling wireless broadband product," if that's faster than 300 Kbps. (That is, if a 2 Mbps service is the bestselling offering, Basic Service stays the same; 3 Mbps, and it gets bumped to 450 Kbps.)
The Premium service level is 1 Mbps symmetrical, at least; no fees are defined. In other coverage, a $21.95 rate is attached for EarthLink's retail price, but I don't see the source. A Digital Inclusion service must be offered to the city for up to 3,200 people at $12.95 per month, and include a CPE that costs no more than $100 and can cost less. Both Occasional Use and Roaming are mentioned with not much detail.
EarthLink can offer a 3 Mpbs point-to-point service when it launches other services, but they can sell it as best effort instead of with a service-level agreement.
San Francisco will consider EarthLink as a telecom vendor for municipal needs, but EarthLink lacks a privileged position in that regard, as is defined in some cities' RFPs or final agreements with providers.
Breaking News! San Francisco has achieved an agreement for citywide Wi-Fi: Nobody said getting Wi-Fi into San Francisco would be easy. Some predicted that negotiation would take forever. If forever is 10 months, they were right. My sources say that an agreement has been reached; more details are forthcoming later today.
EarthLink confirms: "Today, EarthLink and the City of San Francisco reached agreement on the
terms of the contract to build a citywide wireless network." There's no detail in their statement as to the final arrangements for what's free and what's for fee.
Succinct, accurate survey in Time magazine of what municipal Wi-Fi is, the players, its future: An impressive effort in a mainstream publication to explain why cities have rushed in to build or find partners for city-wide Wi-Fi. The writer did an admirable job in providing opinions across a gamut without including, frankly, stupid or uninformed ideas. If you read my site, you already know everything that's in this article.
Time addresses an increasing point of contention that was first raised in 2004, but mostly papered over since: If the city government is involved in running or authorizing a network, to what extent does government have control over and access to your Internet usage; and to what extent is a municipality responsible for what passes over the network. Culver City is mentioned, home of the porn filter and the illegal download (by whose definition, we don't know) filter.
The WiMax developer and licensed spectrum holder NextWave will acquire metro-scale Wi-Fi gear maker Go Networks: Go's equipment uses MIMO to fill more space at lower cost, the company has said. NextWave, a successor firm to the 1990s cell operator and Supreme Court case victor over spectrum auctions, has a portfolio of WiMax hardware and a set of licenses they purchased in the recent advanced wireless services auction. NextWave also acquired some German WiMax licenses last month.
NextWave is obviously assembling a set of technologies that they can roll out in test markets that they have spectrum in, and will be well positioned to test the effectiveness of MIMO-based Wi-Fi as a complement and supplement to WiMax in urban areas.
The Go Networks' deal is valued at $13.3m with a separate assumption of $7.5m in debt. Go would also receive $25.7m in stock for meeting milestones 18 months after the deal closes.
Wired News published this interesting round up of beaches with Wi-Fi: Florida's Haulover Beach is perhaps the only clothing-optional beach that features Internet access. Sans clothes, sans wires: It's a theme.
First Wi-Fi, now Bluetooth: On the heels of Australian technology agency CSIRO winning a key patent suit against Buffalo Technology over the use of very specific elements of newer Wi-Fi standards, the University of Washington's patent-licensing arm has sued four electronics makers that incorporate Bluetooth chips made by CSR into their products: Nokia, Samsung, and both Matsushita and its subsidiary Panasonic of North America. The Washington Research Foundation has patents created by an undergraduate and assigned to the school that the WRF claims are infringed by CSR chips. WRF has a licensing agreement in place with CSR's competitor, Broadcom.
CSR stated today that the claims are without merit. One article says that WRF sued customers because CSR sells chips worldwide, but the customers deliver products specific to the US market that incorporate CSR chips.
The patents were apparently developed in the mid-1990s, according to The Seattle Times, but one of the patents in dispute was not filed until 2003; it was granted in Oct. 2006. The Bluetooth SIG's members agreed to cross-license technology, but WRF is outside that process.
The City by the Bay is "seconds away" from a deal with EarthLink, Google: Daily meetings are finally nearing completion. Negotiation has been underway for 10 months (not daily until recently). But even when the Mayor's representatives have hammered out a contract, the San Francisco Board of Supervisors still has to approve the contract. Some Supervisors are opposed to private ownership of the Wi-Fi network.
The head of EarthLink had taken a leave of absence just a few weeks ago for treatment of a "serious form of cancer": Garry Betty had run EarthLink since 1996, when he took over several roles from founder Sky Dayton, who remained chairman. Betty kept EarthLink a vital ISP, growing subscribers from 500,000 to 5m over his tenure, while other firms collapsed under the weight of competition first from AOL and MSN and then from incumbent telecom and cable firms.
EarthLink has also been the largest player in city-wide Wi-Fi, a move Betty directed as the company's traditional dial-up customer base shrunk, and court and regulatory decisions made wireline resale of DSL and cable access ever more difficult.
My condolences to Betty's family, friends, and his colleagues at EarthLink.
Avis will apparently offer a cell-to-Wi-Fi bridge to car renters: The Autonet Mobile bridge, which receives signals from a 3G cellular network and routes them to a Wi-Fi gateway, isn't unique. Kyocera, Linksys, and Junxion, among others, sell such devices; the former two firms aim at consumers and small businesses, while the latter looks to corporate and fleet deployments. But the deal with Avis would be unique. The bridge would be provided with an DC auto adapter for use in the car.
The New York Times reports that while the deal isn't yet announced, pricing would be $11 per day. The company will also sell its device separately to what it envisions as a mini-van or SUV crowd--parents and kids traveling together--for $400 with a $50 per month service fee. There's no mention of how much bandwidth is included with that fee, which is $10 less than comparable retail 3G offerings from Cingular, Sprint Nextel, and Verizon, which require a voice plan and a two-year commitment.
It's hard to pick the kind of road warrior that would pay $11 per day for the service but wouldn't otherwise already have a cellular data connection that they would use. Traveling groups might be more inclined because of the ease of sharing, but then they might already own some other cellular bridge. However, because it's essentially a Wi-Fi hotspot without any cellular configuration, it might appeal to users who otherwise would find a cell network's configuration daunting. And to those who lack a PC Card slot on their laptop.
While the article notes the legality of Wi-Fi'ing while traveling--"Avis would require rents to agree not to hold it liable for accidents resulting from irresponsible use"--it doesn't mention that the cell operators have a varying amount of interest in allowing bridged Internet connections. Cingular and Sprint Nextel resell some bridges themselves, and their resellers offer bridges, too. Verizon has stated that they don't support their use on their network at all.