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.
Apple jumps out in front, as is typical, and adds Bluetooth 2.0 to its laptops: The entire PowerBook line of laptops will include both AirPort Extreme (802.11g) and Bluetooth 2.0, which adds a 3 Mbps flavor to Bluetooth that an Apple executive told me in an interview today should make it possible for the highest-quality music streaming to Bluetooth headphones among other devices that they expect will incorporate the new technology. This new flavor of Bluetooth is called Enhanced Data Rate (EDR).
Apple says it's the first to add Bluetooth 2.0, and at first glance that seems correct. It's easier for Apple to change the plumbing to add specific hardware to their monolithic operating system than, for instance, for Microsoft to suddenly support 100 different manufacturers' flavors of Bluetooth 2.0. This lets Apple be strangely nimble.
The underlying chips appear to be from CSR, which had the first certified 2.0 chips late last year. CSR will integrate this technology into a new Bluetooth optional card for Dell laptops, the TrueMobile 350 Module. The 300, currently sold as an add-on, supports up to Bluetooth 1.2.
A cautionary tale of using hotspot guides by providers in Paderborn, Germany: For those of you who read German, a very funny short tale of someone visiting Paderborn for family reasons who, armed with a list of T-Mobile Germany hotspot locations, comes up empty again and again (and again).
With consumer 802.11g costing below $70 for good gear, enterprise APs still run you more than $400 with discounts: There is, of course, a price premium you pay for devices that handle VLAN switching, multiple broadcast SSIDs, and other enterprise-related features. But the difference between the underlying silicon is pretty small (or non-existent) these days. You're paying a large differential for brand, service, support, integration, firmware, and the firmware's hooks for management. And switch AP cost even more but provide more flexibility, which should lessen server-room and management costs.
Someone may be using illegal 5 GHz equipment near a Florida air base: A reader sent in a link to this Associated Press story and assumed--as one might--that the military got it wrong and was making literally a federal case out of legitimate use. But the band cited by the Eglin Air Force Base spokesperson, 5.6 to 5.8 GHz, contains part of the range opened by the military through the FCC in Nov. 2003 with some provisos. Someone might be misusing it.
Any device using the 5.250-5.350 GHz band (available for a while) or the 5.470-5.725 GHz band (new as of Nov. 2003) must avoid stepping on existing signals and back down power to only use as strong a signal as is needed at any given time.
It's possible that someone is using a 5 GHz channel that was previously assigned in the 5.725-5.825 GHz band. Since those were intended for outdoor point-to-point use and don't require the limits that are defined for two of the three lower bands, it might be legal and unintentional. Or they're using equipment not approved for use in the US or have modified US-licensed equipment. Or, even, it's very sloppy out-of-band spillover.
And I mean tackles! Pins it to the mat: This article lays out the land in a way that I appreciate: WiMax is an incremental enabling technology not a radical shift in view. No WiMax equipment has been sold yet. None will be sold for at least six months. When it does--and pre-WiMax turns into true WiMax--customer premises equipment will still be pretty steep compared to commodity devices available today.
I had a long talk with SkyPilot the other day, which uses 802.11a-like technology to offer fairly good broadband speeds across long distances. Their tech is totally commoditized. Their CPE cost is $349--for a single unit. It goes down quite a lot (they wouldn't say how much) in quantity. They're about to announce some big customers for their production gear.
WiMax isn't about whether broadband wireless is a viable service to offer. It certainly is. There's no question about that. It's whether a particular instanciation of that technology has any bearing on the deployment unless is has particular advantages that make something possible that wasn't. (That's part of the issue with early MIMO gear for the home, too.)
As I read this Economist article, the real issue isn't whether a company like Qwest would choose SkyPilot's 802.11 over Alvarion's pre- or post-certified WiMax. Rather it's whether "plenty good enough today for real deployment" trumps "much better but much more expensive in the future until we deploy a lot of it."
WiMax has a huge array of benefits for carriers that want to roll out WiMax in the same way they deployed DSL: few truckrolls (because of good non line of sight protocols) and lots of ratcheting in bandwidth offered to provide discrete services that mimic DSL and cable modems. These benefits are more appealing to carriers that are trying to integrate broadband wireless into an existing portfolio. These carriers are also in a better position to bundle applications on top of WiMax thus making it more reasonable for them to eat or subsidize a $500 CPE cost than even a large regional ISP or municipality.
WiMax might be the flavor that telcos and related firms opt for because of consistency, standardization, and technical features. But it doesn't mean that potentially billions of dollars of other gear might not be sold in the meantime that has a very similar function and utility for the non-operator market.
The article also walks through mobile WiMax, which hasn't been finalized yet and is possibly due in late 2006. The Economist points out Intel's previous failures to deliver on anything like a timetable, so 2007 might be optimistic, still.
I admit that I like the idea of mobile WiMax, but I have a hard time believing that it will seem like a good idea when it's actually ready to be deployed. With the increase in speed and sophisticated of systems based on or parallel to Wi-Fi and 3G cellular, it's just hard to see mobile WiMax's place in that ecosystem by the time 2007 rolls around. [link via TechDirt]
Apologies on multiple blank messages: We had a list glitch as we transition to a slightly revised mailing system that will handle lists for multiple blogs...you heard me, multiple blogs. More news soon, and my apologies for shipping you four empty messages.
Peter Judge at TechWorld reports that BT may be introducing a combined Wi-Fi/cellular service, but it's not what you think: When customers make calls in their homes using a combined GSM/Wi-Fi phone, the call is carried over Wi-Fi between the phone and the access point. But the access point is backhauled over the GSM network. Instead of realizing the cost savings of carrying a call over IP over a broadband fixed connection, BT chooses to use the more expensive GSM network. The service becomes, in essence, a method for improving cell phone coverage in the home.
The idea doesn't make sense in any context, but you might understand it if the offering was being made by a cell phone operator that stands to benefit by keeping calls on its network. But this is BT, which doesn't have a cell phone network and will supply the cellular link via resale agreements with a mobile operator. It would make so much more sense for BT to backhaul the access points using its own wired network, using voice over IP and charging customers slightly reduced rates than the cell phone networks for the calls that use the system. That sounds like it could be a profitable service and would allow BT to beat out cellular operators that don't own their own landline networks.
This service is basically an update to the previously announced Bluephone initiative and shouldn't be available until 2006. Given the track record of the Bluephone plan, which was initially set to be introduced using Bluetooth instead of Wi-Fi in 2003, it might not be surprising if the whole plan changes again.
Mobile operator SK Telecom and Earthlink Networks launch 3G/Wi-Fi combo firm: Sky Dayton is the CEO of SK-Earthlink, a new MVNO (Mobile Virtual Network Operator) that will sign up end users with the promise of integrated Wi-Fi and third-generation cellular data networks for voice and data through a single handset. Dayton founded Earthlink and Boingo Wireless.
SK Telecom provides service to 18 million customers in South Korea, and their sister company in the SK Group, SK Teletech, makes advanced handsets for domestic use and internationally to China, Israel, Kazakhstan, and Taiwan. Dayton said in an interview today that SK Telecom offers a remarkable array of handsets including units with hard drives, five megapixel cameras, and satellite television broadcast receivers. "Not all those things are going to be applicable in a US market," Dayton said.
The service will attempt to leverage the growing availability of 3G networks with the increased ubiquity of Wi-Fi hotspots. Dayton says that although handsets and service plans won't be available until as late as the end of 2005, that generally SK-Earthlink would focus on allowing its customers to make voice calls from home Wi-Fi networks, Wi-Fi hotspots, and 3G cellular networks.
They would also leverage Wi-Fi alongside 3G in offering advanced data and entertainment services, although he did not mention specifics. "We think there's a big group of Internet savvy early adopter" consumers in the US who want this but can't get it, he said. SK-Earthlink may use Boingo's platform as the base on which they develop hotspot relationships--Dayton will have an arm's length involvement in that decision--but Boingo will certainly also continue to sell to other carriers regardless.
Other reports note that SK-Earthlink will use CDMA technology, SK Telecom's flavor, which ties them closely in the U.S. to Verizon Wireless and Sprint PCS, the only two remaining CDMA carriers here. Verizon has already deployed a significant portion of their ambitious EVDO network offering speeds of hundreds of kilobits per second, and Sprint PCS is committed to higher-speed offerings than its current 1xRTT rate that compares to modem speeds.
At the same time, Verizon Wireless said recently that 3G will trump Wi-Fi, a sentiment that baffles Dayton. "Wi-Fi is an unstoppable force," he said. "It's like trying to deny there's a big elephant in the room." Both technologies "are great individually, but if you put them together" they allow the best connection wherever a user is, he said. Dayton's comment seems particularly apt on a day when Strategy Analytics released a report showing that Wi-Fi will significantly reduce 3G profits.
Dayton's involvement in an MVNO is characteristic of his previous businesses. An MVNO has no infrastructure but purchases access from existing carriers for resale. "There's no political problem in a new MVNO: we don't have to build new networks," Dayton said. Earthlink was founded as an ISP with no modems. Boingo sells hotspot service without building hotspots. SK-Earthlink is unique for him in supporting end-user hardware: the handsets that will be sold to subscribers.
The 50-50 joint venture will incorporate Earthlink's existing Blackberry, cellular data, and Wi-Fi subscribers. Boingo provides Earthlink's Wi-Fi service through a reseller agreement. Reports indicate Earthlink has about 30,000 customers for these services.
Until today, Sky Dayton served as chairman of Earthlink Networks, a company he founded more than a decade ago as a dial-up Internet service provider with no modems. He will remain on the board, but steps down as chairman to avoid a conflict being the CEO of a joint venture between a firm he chairs and SK Telecom.
Let's compose a little song: AP covers A, B, G, I, and sometimes E: Is it my constant singing of children's songs to a baby that makes me think of the 802.11 working group and Wi-Fi Alliance's alphabet soup in the form of a chant with music? I don't know, but a colleague at Associated Press filed this cogent deciphering of the goop into coherent advice for consumers.
The brief version: buy 802.11g now but not pre-N.
I get the last word in this AP story when I talk about pre-N or early MIMO gear: "You'll be buying equipment that will be obsolete in the near future and will become cheaper in the near future," said Fleishman. And I stand by that remark.
If any manufacturer producing MIMO equipment today is willing to guarantee on the record that they will offer fully compatible final 802.11n upgrades--firmware or even hardware--to the certified Wi-Fi version of that standard for equipment that a consumer buys today, I'll recant, and praise them lavishly, as it's the only way to ensure that consumers don't pursue a near-term dead end that costs more.
No manufacturer is willing to say that what they ship today will work with the highest speeds of 802.11n. There's even a high chance that equipment from different vendors using different chips won't interoperate at speeds above 802.11g using today's gear. You'll still get increased distance and throughput, but unless you have very specific needs today and can use homogeneous equipment, wait, wait, wait, wait.
"Who's the leader of our pack that's here for you or me? Eight oh two eleven g with double-u pee aaaaay!"
Update: I've had a little pushback on this post and my quote in the Associated Press. A few folks have written on forums that my statement closing the AP story applies to all technology. True, but this AP story was written with the consumer in mind, not the early adopter. Consumers who buy MIMO today need to have a compelling reason of range or throughput to opt for MIMO instead of 802.11g with the variety of range extension technology now offered by major chipmakers and consumer Wi-Fi manufacturers.
If you don't need range and speed, then why buy equipment that will be slower and almost certainly more expensive than the approved flavor of 802.11n next year? It's very likely that MIMO equipment sold by the end of 2005 will have dropped greatly in price and will more closely resemble what the final 802.11n spec will look like--there are only two MIMO proposals left to consider by the 802.11n group, which means it's getting clearer what the hardware requirements will be.
Some users need more range in their homes, can articulate that problem, and can attach that to a higher-priced alternative that meets their need. I haven't seen MIMO versus range extension for 802.11g tests, though, and I'd like to know whether any of the chipmakers' extensions are far outstripped by the current crop of MIMO devices. They may be, but with $60 to $80 for an 802.11g gateway and $130 to $150 for MIMO, the difference as to be significant.
Speed is certainly another consideration. MIMO offers greater throughput than any standard or proprietary wireless protocol in use. If you need speed, then it makes sense to buy new adapters and new gateways to achieve the best possible speed. But it's rare that a consumer needs that kind of speed right now; it's more of an early adopter or business concern.
I didn't mean to be fatuous by saying that technology becomes cheaper and obsolete over time. Rather, I wanted to emphasize that the very reason you would pay a premium for pre-802.11n devices right now will be so rapidly eroded that it's worth waiting. If you buy now, you get double the speed, triple the throughput, and double the cost. But you only get those speed improvements for as long as you choose to stick with the proprietary equipment throughout your network. Consumers do buy and sit on equipment for long periods of time which is why I urge patience for the truly next-generation flavor that maximizes flexibility through interoperability.
Wait about 12 months and you'll get quadruple the speed, as much as sextuple the throughput, and nowhere the cost of today's early MIMO devices--and a piece of equipment that has a future path for its highest speeds.
And you'll be ready for streaming video.
David Haskin at Mobile Pipeline interviews Philadelphia CIO Dianah Neff: The article is wide ranging and reveals more of the plan that will be unveiled Feb. 7. Neff argues--as I have--that it's disingenuous for giant telecoms to decry city tax-free financing because, first of all, that's not how Philadelphia will pay for this network (nor did they ever say they would use tax-free bonds for it), and second, the telecoms have received without complaint billions of dollars in subsidies.
As one article has pointed out, Verizon received enormous payments in the last decade to encourage them to build services that they didn't. Instead of being penalized, the bill passed in Pennsylvania gives them another decade with more incentives.
Neff tips her hand a little about the unique public-private partnership that she's been alluding to in recent weeks: I've been pushing the notion of vendor-neutral municipal networks that provide a place for all ISPs on an equal basis, including Verizon, and that don't put the city in the business of being a provider. Rather, a municipality becomes an enabler; the money for logical access is all spent in the private sector and non-profit sector. Neff cited the existence of hundreds of ISPs that she hopes will be part of the city's venture, which makes it sound an awful lot like a vendor-neutral network.
News from inside says that MITMOT is out of the running: The Motorola-backed 802.11n proposal was the poorest vote getter at the Nov. 2004 IEEE meeting that include Task Group N (high-throughput enhancements for 802.11). The third worst recipient of votes was Qualcomm's proposal; they've joined TGn Sync. TGn Sync and WWiSE will be the remaining proposals considered moving forward.
For the security wonks, Ethereal 0.10.9 now detects and flags weak initialization vectors (IVs) for WEP keys: An initialization vector (IV) is an attempt to increase randomness in a publicly available encryption stream. Combine a truly random IV from a large number space with a key set by a user and each packet has additional protection against brute force attacks. WEP was weak to begin with because the IV space was too small, forcing reuse.
Errors in implementation meant that IVs are rapidly reused on some networks. A flaw in the encryption algorithm further means that certain IVs, called weak IVs, reveal more information about the secret part of the WEP key than others--about 9,000 weak IVs out of 16,000,000 possible ones. (WPA, by the way, has a 48-bit IV instead of a 24-bit one without the weakness problem.)
The weak IVs are much more interesting than others because they have a statistically higher likelihood of resulting in a crack. Gathering weak IVs quickly produces a crack faster than gathering lots of strong IVs. (Some WEP-generating Wi-Fi adapters exclude weak IVs from their IV creation process as part of firmware design.) Gather a few thousand weak IVs, and you have a chance at cracking the key. On busy networks, this might be a few minutes using the latest cracking software.
Ethereal 0.10.9 detects these weak IVs, which either means you know you can crack a WEP key quickly or you know that your network has weak protection and ought to be upgraded. [link via Nigel Ballard; thanks to him and Jim Thompson for reality checks.]
Simple Web requests reveal admin user name, password, WEP key, and SSID: 3Com has fixed the glaring hole as of Jan. 19 based on an iDefense report that was provided to them in advance and released yesterday. 3Com users of the affected equipment are obviously advised to immediately update. It's unclear whether the URLs noted in the report are easily available over the Internet: a quick Google check using "inurl:" (looking for text in an URL) didn't spot any 3Com routers. It's likely that remote administration has to be enabled. [link via Frederick Wamsley, the expert behind Beryllium Sphere]
Updated: The FCC Chairman Michael Powell resigned today effective in March: I have completely disagreed with Powell's support of media consolidation and differed on his involvement in increasing the role of the state in acting as a nanny; I do believe that public airwaves require some oversight and regulation that's non-partisan and objective based on criteria instead of letters written to the FCC.
But this site's myopic focus will remember Powell's legacy more for his unstinting support for the opposite of consolidation in spectrum policy. Over the last four years, Powell by his public statements and, ostensibly, private actions has managed to open more spectrum, consider innovative secondary uses of licensed spectrum, and build a framework for cleaning up the messier and least used bands that are needed for 3G and beyond and WiMax and beyond.
In these areas, Powell's leadership encouraged technologies that aren't centrally owned or controlled and that may, in fact, dislodge primacy of wireline incumbents.
Powell will step down as a commissioner and as chair.
The huge Bainbridge Island-Seattle ferry run gets Wi-Fi: The Washington State Ferry system continues to roll out Wi-Fi on board using a system I wrote about in the New York Times last July. The latest news is that they're ready to move from testing into full-scale free (for now) use on the largest run in the system, Bainbridge Island to Seattle. Bainbridge is a bedroom community of sorts for downtown workers, but has a very active life of its own. That one run, which carries 6.5 million passenger trips per year, is larger than entire ferry systems elsewhere in the U.S. The WSF carries 50 percent of the passenger trips in the U.S.
The Wall Street Journal exhaustively surveys the lay of the land for broadband wireless deployment: I estimate that this article took a few dozen hours and weeks of reporting--some of which surely found its way into other articles--to pull together because of the number of companies and technologies involved. It's a great overview that focuses on WiMax, metro Wi-Fi, and other broadband wireless as an alternative to the wirebase that's needed by cell companies and service providers who don't own copper.
I particularly like the neat turn of phrase that encapsulates the entire WiMax branding and hype problem: One of the technologies drawing the most attention is WiMAX, which is similar to the popular Wi-Fi standard that millions of people have used to set up wireless networks in their homes but is slated to have a range of several miles. Since WiMAX has yet to be certified, companies are using precursors to the technology.
Exactly. Precursors aren't necessarily worse, but they're not interoperable and they don't bring the benefits of mass-market standardization to reduce the CPE (customer premises equipment) that will ultimately make broadband wireless affordable to the average home instead of a subset.
I love this bit of specious reasoning quoted about the landline side of things that are causing this competitive wireless marketplace to emerge. The Bells argue that they shouldn't be forced to share their lines. "We're incurring all of the costs of building these networks and we don't feel we have to share them with our competitors below what it costs us to build and maintain our network," says BellSouth spokesman Jeff Battcher.
I don't think anyone has ever asked the Bells to subsidize the cost, but rather to provide an accounting that shows the true costs. It's clear that because the Bells can bundle services and make money across an entire customer package that they have every motivation to make their wire costs much higher to discourage having to resell access at a price that allows competition. In other words, if the Bell companies can arrange a markup over costs, why can't they resell at wholesale with a margin for competitors? They can, but they want it to appear as though they can't to preserve their bundling profits.
Also, I guess nobody every explained to Mr. Battcher that monopolies, natural or regulated, are subject to different rules than companies competing without any advantage. The Bells own the wire; they should be forced to share unless you believe that consumers should pay the maximum possible price rather than the optimal price decided on by a marketplace. Those focused on business returns and shareholder value would argue the highest price the market will bear is best; those focused on consumer issues might maintain that more competition would produce an ideal price set by the contention of service in the bazaar.
But broadband wireless coupled with pressure from cable operators has at least forced a semblance of competition with much, much more on the way.
What's most important about the survey of the landscape in this article is that it shows how widespread the tests are already by major firms and how many tens of billions will be poured into all forms of broadband wireless in the very near future.
Fellow Glenn, the InstaPundit, joins in on the EVDO debate as a real user: We share a name and a political viewpoint--not in the red versus blue state sense, but around EVDO and Wi-Fi. Glenn brings perspective as a Verizon Wireless EVDO user and as an enthusiastic employer of Wi-Fi as well. Glenn thinks that Wi-Fi is a bottom-up phenomenon that has different characteristics than a top-down cellular network. He also points out good issues about battery life.
If you can't use EVDO for more than a couple of hours without recharging, then you need an electrical outlet. Which means you can probably find Wi-Fi near that electrical outlet.
Comments without real attribution are unlikely to be approved: I've seen a small contingent of comments appear on this blog recently--which requires moderation for those comments to be publicly posted--that accuse me of idiocy for not fully understanding Verizon Wireless's EVDO plans, among other subjects. While I like critique and am willing to allow comments that take an opposite stance to my own, I'm unlikely to ever approve a comment with a fake return address and no name that calls me a moron.
The issue, in part, is that the phenomenon of sock puppets, which I believe was coined in the Usenet newsgroup world. A sock puppet is someone (or some company's) alternate identity that chimes in, "Hey, Bob, what you wrote? I totally agree! You rock!" I'd expand that definition to include posts that take a strong stance that lack any credentials. (Another category: posts that call me an idiot because of typos rather than saying, "You might want to fix this typo.")
Anonymous posts that have something interesting to say that advances a subject through logical analysis have a high chance of being approved. Anonymous posts that have ad hominem attacks and advance a position by repeating it don't. Overall, re-enabling comments on this site has been very useful, and dozens of comments have been added.
One of the great attributes of the blogosphere echo chamber is that if you start your own blog and point to a news site or another blog with a comment, your comment gets incorporated into that item's fabric. I won't tell anyone it's my ball and glove and I'm going home, but, rather, there are an infinite number of balls and gloves and playing fields, and I encourage everyone to have a forum in which they can fully express their opinion without moderation.
I believe I'm seeing sock puppets because there are some sites that have written really scathing or interesting rebuttals to posts here at Wi-Fi Networking News, and I love reading those because there's a real person willing to identify themselves behind it. I might not approve comments for this site that look anything like those other blogs' remarks, but that's the great thing about the infinite forum that is the Internet.
It's the title of a They Might Be Giants song, and also a meme gone round the world: A number of people sent me links to articles about "evil twins" in the last couple of days. These articles are coining a new term for what has been called "soft APs" (software-based access points), a real problem that's been found in the wild for some time. These evil twins use software that creates a rogue access point which has the same name as a nearby network. AirDefense detected a number of these starting during their ongoing scan of the June 2004 Wi-Fi Planet Conference.
Because most operating systems are promiscuous by default, they will join any network with the same name as one they have joined before. If you're warned about joining network "linksys" the first time, you won't be warned the next. And there's nothing that helps you differentiate between a good "linksys" and a bad "linksys." If you have WEP or WPA encryption enabled, however, you won't be able to join an evil network because the key won't match. Public hotspots are really the biggest place to become a victim. (Glynn Taylor wrote in to note that the demonstration of this attack, Airsnarf, also includes a beta of a detection tool that notes whether an access point's characteristics have changed.)
This is why I highly recommend that all users of public networks employ some level of protection for any passwords that may travel across their networks. If you use SSL email client connections for POP, IMAP, and SMTP or an SSL-enabled Webmail site, just for instance, you're secured because an "evil twin" can't provide false digital certificate information to capture those sessions.
Web designers should always use Secure FTP (SSH over FTP), which is an encrypted form of FTP. If you don't know how Secure FTP works, find an ISP that does.
VPNs are now cheap and plentiful for rent. I recommend HotSpotVPN.com all the time because it's cheap ($8.88 per month and cheaper for longer pre-paid periods) and simple, working with software pre-installed on almost all platforms shipped in the last five years. You can also buy and install a Buffalo secure Wi-Fi gateway on your home or office network that offers full VPN protection for a small office for less than $200.
The 802.1X standard also alleviates this problem. If you log in over 802.1X, you'll be warned if you can't authenticate to a network. There are potential man-in-the-middle attacks, but properly monitoring certificate warnings works there, too. For instance, if I try to connect to my office AP, for which I've already accepted and installed a digital certificate confirming its identity, and an "evil twin" gets in the way, my 802.1X client warns me.
This is one reason, I'm sure, that T-Mobile was so eager to roll out 802.1X on their networks. Their client software has the root authority for their 802.1X service preinstalled for out-of-band trust that allows you to reliably only connect to their networks. Anyone trying to spoof a T-Mobile 802.1X-enabled AP won't get far.
One factor holding back public hotspot 802.1X deployment is that many hotspots use inexpensive access points that lack (or used to lack) the ability to operate discrete VLANs coupled with separate broadcast SSIDs. What this means is that T-Mobile can operate two logical networks--one protected by 802.1X and the other with a gateway page--without having to install two pieces of hardware. That was a missing piece that's now available, and this evil twin problem is practically a call to arms to hotspot operators to take a stand and start an 802.1X migration for their customers' benefit.
All this to say that we're about to see a dramatic acceleration in authentication and encryption that will bypass the utility of evil twins. The biggest factor holding us back? A lack of free legacy 802.1X clients for Windows 98 and Me, as well as flavors of Mac OS X and Linux. You can purchase clients for most older operating systems from companies like Funk and Meetinghouse, but because only Windows XP and Mac OS X 10.3 have built-in 802.1X mean that we have a migration ahead rather than a simple switchover.
The wait is over and Centrino's latest revision ships: the new version offers both 802.11a/g and 802.11g options. When laptops based on the newer chips ship they'll include an updated wireless networking program, but I don't have details on what looks like. Centrino's wireless app was interesting because it used a combination of graphics, spatial relationships, and text to help set up and troubleshoot Wi-Fi connections.
Panera will increase to 600 free Wi-Fi enabled restaurants: The company currently reports 490 of its stores have Wi-Fi. Today's announcement that Savvis will handle the networking says they'll offer it in 600 outlets. The company has over 700 stores.
SBC FreedomLink will unwire California state parks: Ack, we'll never escape the Wi-Fi menace. Just try to find some nature without electromagnetism. But, more seriously, people who spend time traveling or need a link to home or office now have a less invasive method than cell phone calls. Maybe we don't need to send email to say, "It's beautiful here," but human nature is about communication.
The first park unwired will be outside of San Diego, San Elijo Beach Park (pictured), and 84 other parks follow during the next six months. Interestingly, the cost for a day's use is $7.95 for non-subscribers versus $4.95 in other venues. The parks will offer free access to state-operated Web sites, which might be a real treat when you're trying to find a campground and can use their online reservation system or check weather and roads.
California has 700 SBC hotspots, according to the press release. SBC users have access to well over 6,000 locations through Wayport managed services (UPS Store and others) and Wayport roaming (McDonald's, hotels).
For a dissenting view on park-Fi, dig this camper's comment: "We'll stay out for three months, and I manage just fine without it," said camper Gypsy Donachy.
[Photo courtesy of Phil Baker out of his window.]
Perhaps we'll have a story every day on Philadelphia: Maybe it requires its own Philly-Fi Blog? The city's CIO Dianah Neff spoke at an MIT conference on Tuesday and provided a number of specifics that, if reporters had been paying attention, she had said in other public forums over the last couple of weeks; those statements received coverage, then, too.
Icomera has signed SJ for 85 trains with 3G/Wi-Fi access: SJ is an enormous rail company covering parts of Scandinavia, including Denmark, Sweden, and Norway, with links out of country as well. Their Intercity and Commuter lines comprise 85 trains and they'll all have service by summer. This is the first widespread train rollout in the world, with trials or single lines connected via 3G or satellite uplink distributed via Wi-Fi onboard. SJ carries 70,000 people a year, but that must be actual people, not trips--that's the first transportation company I've seen that's underplayed its numbers.
Update: Fortunately, the 70,000 figure is a translation error of some kind. A few folks of the Scandinavian variety wrote in to point out that the company carries 70,000 passengers a day. That's more like it. I was picturing quite empty trains.
Verizon will sell a cordless phone with DSL in the base: I lament the long-lost HomeRF, which integrated data, streaming media, and telephony long before its time, but died because of a regulatory issue, speed, and clarity in marketing early products. The Westell phone system that Verizon has licensed incorporates a DSL modem into the phone's base and uses 5.8 GHz for voice transit; it's also an 802.11g Wi-Fi hub. The phone has a display that offers primitive contact, calendar, and information services.
Verizon first discussed Verizon One in early 2003, then showed a demonstration at an investor's event in Jan. 2004. The photo at upper right is from that event, and may or may not look like the Westell product: there are no product shots available yet. The product works with Verizon's iobi, too.
Now if Verizon would follow Verizon Wireless's lead, they wouldn't allow regular VoIP over the Wi-Fi network on this system, because nobody ever promised that Wi-Fi that you bought as part of a home networking system would be free to use for any purpose, right? Wait, I'm channeling a Motorola spokesperson.
Wi-Fi use exceeds other methods in the home, says Parks Associates: Wi-Fi beats Ethernet in the home in the U.S. with 52 percent using Wi-Fi and 50 percent using Ethernet. In Canada, it's 32 percent Wi-Fi, 43 percent Ethernet, and 26 percent don't know. (More than one method can be used, and often is.)
The WiMax Forum has postponed the start of plugfest, the time when interoperability tests were to be done in the initial steps toward the equipment certification process: Plugfest was supposed to start this month but now has been tentatively set for as late as June or July. The setback would appear to add half a year to the certification process.
While few expected the schedule to go exactly as planned, six months is quite a long delay. In the meantime, vendors will continue to market their "pre-WiMax" equipment because they're now put in an awkward position. They can't exactly hold off on selling gear while they wait for the process to move forward. Also, with all the hype around WiMax, they'll feel compelled to use the term in their marketing efforts or risk losing potential sales.
The delay itself also risks being overhyped. WiMax is getting a bit of criticism of late, as 3G networks launch, Wi-Fi coverage expands and questions surrounding the demand for a fixed WiMax service surface. Realistically, practically every standards process has taken longer than expected.
It's a bit ironic that Intel is being blamed in part for the delay because it has yet to release its silicon which will be used in the customer premise equipment. It doesn't look great for WiMax's biggest cheerleader to be partly responsible for the delay.
MIMO alliances are shifting as Qualcomm brings its good self and patents to TGn Sync: EE Times reports that Qualcomm has dropped its own plan, a kind of third alternative, and will join up with TGn Sync to support 20 companies including Cisco, Intel, and Atheros. The issue of RAND is not mentioned in the article: that's reasonable and non-discriminatory terms for licensing intellectual property.
The TGn Sync Web site says their proposal offers up to 315 Mbps of raw bandwidth with two antennas. Throughput would likely be at least 200 Mbps based on other details they provide. TGn Sync wants the option for 10, 20, and 40 MHz channels and two to four antennas. (The article cites different numbers, for some reason.) Lower speeds would be possible with fewer antennas and less spectrum allowing MIMO to be fully deployed in newer devices that don't have the power or necessity for full TGn Sync speeds. They even expect 600 Mbps flavors in what they describe as "larger" devices.
The folks at WWiSE, which includes Broadcom and Texas Instruments, are concerned about international use of 802.11n, and propose four antennas and the current 20 MHz bands. This provides speeds up to a raw rate of 135 Mbps. The article doesn't mention WWiSE is interested in options that would allow 2, 3, or 4 antennas and 40 MHz bands. WWiSE states in their primary navigation that they're offering RAND. The technical details at WWiSE show their position clearly: they believe they're offering better spectral efficiency.
Cell subscribers sue Verizon Wireless for disabling Bluetooth file transfer: There was word all over the Net that this was brewing, and this report says the suit was filed. Verizon Wireless disabled Bluetooth file transfer (and possibly other features) on its Motorola v710 phone. The suit alleges, as some online mobile folks have said since this came to light, that Verizon disabled the feature to force its subscribers to transfer photos only through its higher-priced data service offerings.
Bluetooth operates at about 700 Kbps of real throughput; Verizon's EVDO network, as cool as it is, can only handle 50 to 100 Kbps upload speeds; slower where there's only 1xRTT available, too. To transfer photos to your computer, you'd need to subscribe to a Verizon data plan and photo plan, transfer the photos, and then download them. So it's a three machine process.
The suit may hinge over whether Verizon Wireless misled customers, which it appears prima facie that they did not. As a Motorola spokesperson said, quite amusingly to my ear, "Nobody in the industry has ever said that Bluetooth would always be cost free. It will vary from operator to operator."
It's amusing because it implies that files stored on your phone don't belong to you. It's akin to the increasingly common argument made by companies that design devices to play or store media that you purchase or create that the content that you own or have rights to use doesn't really belong to you. What next? Will Verizon Wireless invoke the DMCA in its defense? (Don't ask me how, but remember Lexmark and their DMCA printer cartridge suit.)
Waikiki, Hawaii, is a hotbed of hotspots in paradise: But roaming issues are a sticking point for users, networks, and the local accommodation association. Roaming may be less of an issue across the Lower 48, where networks sit cheek to jowl in some urban areas and are spread far apart in others: there's more flexibility to choose networks that have the best plan and fit into your budget and travel plans.
Waikiki is developing what seems like an overabundance of networks, none of which--according to this article--have roaming agreements. That's annoying and confusing to visitors, and it doesn't truly help grow the networks' use which is what the hotspot operators seem to be focused on: roaming increases use by decreasing cost per use. It's how the Internet grew, folks. More points of access at lower cost equals more usage thus repaying a network's costs more quickly than an island (pun intended) of access. There's an adjustment period where high daily rates aren't offset by more use, but users get retrained and it starts to balance out.
There's also the free versus fee issue: a Seattle visitor chose his hotel for free Wi-Fi over two with $10 per day Internet rates. The article doesn't mention the room rate: was the free hotel also cheaper in that regard? It often is.
It's a side note in an article about the venerable firm's business: But they're trying Wi-Fi in a few shops in Chicago. Starbucks, they're not, but they offer cheap coffee (which is generally pretty good in my experience across the country) and piping hot fresh donuts without the thousand pounds of extra sugar (just a few hundred) layered on Krispy Kremes. They even taste good cold.
Dunkin' Donuts would be an interesting addition to the Wi-Fi landscape because their locations tend to be in my travels in parts of towns in which there is less of an accumulation of the kinds of businesses that already provide wireless access.
Interesting editorial concludes that spectrum management is leading to municipal wireless fights: If you follow Thomas Hazlett's logic from start to finish, he argues that the reason there's a pull for municipal wireless--including from municipalities that have fought or imposed tough conditions on other forms of broadband in the past--is that companies that would like to provide such services are stymied by bandwidth scarcity. If they had the bandwidth, they'd have built the service, and would already be providing effective competition for incumbents. Craig McCaw provides one example: he bought MMDS bandwidth and now is looking at providing competitive offerings to wireline broadband.
I agree with elements of his reasoning. If the cell companies and other firms interesting in wireless data had enormously more bandwidth, it would be substantially easier for them to trial and roll out data services without worrying about scavenging bits and pieces and trying the most advanced, most spectrum-scrounging alternatives. (Scarcity promotes ingenuity, too, so there's a case to be made for giving billion-dollar companies a box full of parts and asking them to build a spaceship.)
Hazlett is dead on about the lack of coordinated spectrum policy management in the U.S. It bites us again and again compared to the lockstep of parts of the rest of the world, making us less competitive in that businesses have to support many standards and many bands. There's no logical path forward. Diversity breeds strength, one might argue here, too, but sometimes easier is just better. The efforts introduced by FCC chair Michael Powell to reform the MMDS band, thus reallocating and freeing up massive amounts of sweet-spot spectrum plus the eventual freeing of VHF and more UHF bands when the DTV transition is complete (in some future decade) might eventually provide the necessary bandwidth for "easier."
Hazlett does sidestep the entire cost issue. 3G spectrum was available in Europe, and telcos practically bankrupted themselves bidding it up during the bubble's growth worldwide. American carriers avoided spending resources just before a time they could scarce afford it, and that might be why we have six strong carriers who just collapsed into five and possibly soon four.
UWB is a great idea for a cordless phone replacement: The chance of interference is about nil, and while UWB needs short distances for high speeds, it can actually maintain decent speeds--certainly high enough for voice codecs--at much longer distances, such as throughout a typical modern home. If you already have a UWB hub or infrastructure for home entertainment, this isn't that weird an idea. ZigBee, 802.11n, and other technologies may bite deeply into parts of home entertainment management and streaming, however, before UWB is widely enough deployed. (Why do you think MIMO is so big, so fast? Streaming video, folks, streaming video.)
Communications Daily reports that Philly will announce network details: The article recounts the bumps along the road and doesn't clarify what "announce" means in this context. The CIO of Philly, Dianah Neff, is quoted saying that an initial $10 million investment will meet their goals for providing a mix of free, subsidized, and commercial access, and could generate $5 million in unattached cash flow (sort of like profit) after just two years.
Intel will help municipalities fight pro-incumbent, anti-municipal services legislation: Intel Telecommunications Group head Sean Maloney says the adversarial role between municipalities and incumbents communications providers is counterproductive, and the laws that require right of first refusal are not a good model. He doesn't like free Wi-Fi, either, though, because it undercuts that there profit motive we've heard so much about.
Maloney notes a variant on my take on the subject: I believe municipalities should focus on hiring a company to build a vendor-neutral network over which any service provider can offer their logical signal on a level playing field. Maloney wants municipalities to create a bidding process in which incumbents have just as much reason to build the infrastructure as they would on their own.
Intel needs muni-Fi in order to push mobile and fixed WiMax. An Intel spokesperson tut-tutted about lobbying, but said Intel would use its influence. Which is, ah, let's see. Well, close enough, but we'll be happy to watch King Kong fight Mothra from a distance.
Not good news for those who thought Verizon Wireless was clueful: This New York Times article does a terrific job explaining both technology and applications of those who want a primer in 2G, 2.5G, and 3G cellular data networking, even though the author doesn't like the term 2.5G and mentions it in passing, which then sort of confuses the role that EDGE plays in the transition. (The technology went 2G, 3G, 2.5G: 2.5G was a bridge inserted to make sure the cell companies had some faster speeds before 3G could be deployed.)
But the doozy in this piece is the quote from Verizon Wireless's chief marketing officer--note that word marketing--"For the business customer, especially the laptop guy, it's all about speed and ubiquity," Mr. Stratton added. "I think this really puts a hurt on the entire Wi-Fi concept for the business user."
He's made the classic mistake of believing his own advertising copy and has the sound of a company without a Wi-Fi plan. EVDO is maybe 300 Kbps on average on a good day. As additional users with unlimited plans and using Verizon's new video delivery service start crowding on urban networks, available bandwidth will decrease, although average speeds shouldn't drop too far; it's more likely burst speeds that jump up to 1 Mbps disappear, from my understanding.
Wi-Fi is constrained by the back-end pipe. Right now, that's 1 to 1.5 Mbps in most locations that are serious about Wi-Fi, which would include all the major domestic hotspot networks, comprising well over 15,000 of the hotspots in the U.S. That bandwidth will climb without extraordinary additional costs. Covad can provide 6 Mpbs down/1 Mbps up ADSL in parts of the country for about $80 per month. EVDO speeds and availability on cells is constrained by spectrum and technology; higher speeds are years away per earlier articles on this site.
As applications increasingly become bandwidth dependent--podcasting and video delivery being too leading-edge trends--the user who now might be content to spend $80 per month for ubiquitous EVDO in many major cities and can stand to wait a few minutes longer to download his or her PowerPoint presentation, well, that same user signs up for Cingular's future UMTS network along with a FreedomLink unlimited Wi-Fi plan, and, by the way, uses VoIP at home and the on road to cap long distance expenses and be reachable.
If Verizon is really looking at EVDO as a single mode delivery mechanism over which they deliver a variety of services, they're out of step with what SBC (and Cingular, as a majority-owned partner) is telling the industry is the future: integration across DSL, Wi-FI, and cellular, with applications layered across all three modes of delivery to their customers. Customers seek the right kind of bandwidth for the application rather than stapling the application on top of the bandwidth that the firm has available.
Among other trends, Verizon is missing the VoIP train and the increasing trend for bandwidth heavy and low latency services, and those applications could trump video on a tiny screen wherever you want it.
SecureEasySetup offers one-push security, but does it deter malicious attacks? A few days ago, I wrote about Broadcom's new SecureEasySetup security offering, which allows a home user to setup and configure WPA Personal by pressing a button (software or hardware) on an access point and any client device, whether a computer adapter or a consumer-electronics device. The system is pretty slick and has a great degree of security and ease of use.
But in contrasting SES with a similar offering from Atheros, JumpStart for Wireless, both of which were announced during CES last week, I noticed what I thought was a gap in the security during a period of contention for SES. Atheros's system requires that a user enter a password when setting up security, but the password is out of band--not revealed over an insecure network--thus allowing a simple password to create and distribute a robust WPA encryption key. (For more details on both systems, consult this in-depth article contrasting the two.)
SES seems to lack true out-of-band security. That is, once the button is pressed to initiate a key exchange on an access point, any client within range can jump in and receive the key if they slip in the queue during the two-minute period of contention before the user pushes the button or initiates SES on the client device they want to add. (After two minutes, the button needs to be pushed on the AP again to restart SES; Broadcom estimates most users will start the key exchange in a matter of a few seconds on the client device.)
I spoke today with David Cohen, who has been deeply involved with WPA at the Wi-Fi Alliance in his role as senior product marketing manager at Broadcom. Cohen noted that there are methods by which an out-of-band element is layered on top of SES to prevent malicious clients or putative cracker software from becoming valid clients on a network in which security keys are managed by SES.
The first is that client devices will be expected to provide a warning when another client slips in the queue during the two minutes from when the access point is activated to watch for a client's request before the desired client. For instance, if I push the AP button, walk over to my computer, and click a software button, the Broadcom system includes a monitoring element that will allow the client software to signal that my client didn't successfully gain a key because another client gained access. "You'd have to be realistically under deliberate attack and then you'd get notification," Cohen said.
On a consumer-electronics device, Cohen said the device could either signal this problem through a display--such as an LCD or TV--or for non-display devices it could flash a red LED rapidly. Cohen emphasized that Broadcom is providing a framework and that individual manufacturers will choose how to integrate SES into their products, including this sort of feedback to the user. But Broadcom considers this notification significant because it removes edge conditions for security.
In this case, if a user then wanted to change the key for their network, they would hold down the access point SES button for at least five seconds or choose an option via software for APs without a hardware button. This triggers the AP to generate a new key, and now the user can re-establish keys for their devices. If the problem recurs, there's a second option for client devices that have both Ethernet and Wi-Fi interfaces.
The second option provides total out-of-band security: you can operate SES over a wired network. Plug your laptop into an Ethernet network or a port on the access point, push a button on the AP, and now push a button on a client. SES handles the transaction out of the Wi-Fi band--in the wired band, which is ostensibly far more secure on a home network--and then the device can be used on the Wi-Fi network. (Atheros's JumpStart requires that the AP is configured with a password via an Ethernet network initially, which allows clients to join over Wi-Fi using a password that was protected out of band.)
Cohen emphasized that my scenario of a malicious attack--and accidental one is extremely unlikely--is an edge case. "The global problem we're trying to solve is over 80 percent of the networks out there are wide open," Cohen said. "Hackers are going to jump on those open networks. We want to bring that number down."
Cohen's response to my concern answers a fairly wide swath of my issues about a lack of out-of-band confirmation, but it still leaves open what is more realistically a denial-of-service (DoS) attack against SES rather than a security hole. Because of the notification element that's part of SES, a malicious client won't go undetected. But it could make it impossible for a user to set up SES. (They can always fall back on manual WPA setup at that point, of course, or employ the "nuclear" trigger: one finger on the AP button and one finger on the client button.)
If malicious client software were written so that it constantly scanned for SES transactions and instantly leapt on any AP that's looking for a partner to dance with, this could block a user's ability to ever set their network up with SES. But I can see a small improvement that would prevent this. SES could cause an AP, after having SES reset to produce a new key, to reject the last successful SES transaction's MAC address or other radio and network characteristics.
In that scenario, the rogue client may be constantly scanning, but it's locked out during the next cycle--or even permanently. (Of course, MAC addresses can be spoofed or generated randomly, so I may not be thinking big enough.) Because this would only be triggered when a user reset the SES key on the AP and only after two clients contended and one lost, this wouldn't be trigger in normal configuration circumstances. This would be a firmware change, so it's doesn't require a radical rethink on the part of Broadcom or its partners.
SES is almost all the way there, and it does still have an advantage of ease of use over Atheros's JumpStart. JumpStart clearly is trying to combine the best of out-of-band security while still using the same band for securing a network, but it's interface issues--having to enter a password--definitely bumps its complexity higher. Broadcom would like to keep the margin of attack small, but if the wrong software were to appear, users might find themselves slightly stymied and turned to manufacturers for help in sorting out just what's happening on their network.
The Wall Street Journal reports Cisco and Airespace have agreed on terms: The deal, first indicated by News.com days ago, values Airespace at $450 million. The company makes a switched wireless LAN product line that leaves radio frequency intelligence in the access point and moves intelligence into the switch. Cisco's product line has smart APs which track the user AP by AP instead of across a switch. Cisco confirmed the deal with a press release this afternoon.
The deal will allow Airespace to solve a major problem affecting WLAN switch vendors, only partially solved by them introducing their own Layer 2 switches: dealing with the bottlenecks of pumping traffic across their centralized switches from increasingly speedy and loaded WLANs. With Cisco's expertise in the increasingly common 10 Gbps Ethernet switches and backbones, this should allow Airespace to more easily extend their intelligence without bogging down networks.
This becomes especially important when networks speed to 100 or even 400 Mbps on the edges. With just one or two gigabyte Ethernet ports on a switch, it doesn't matter how smart you are: you just run out of room.
Airespace has a reported total of $58 million of investment dollars, making a nice return on the deal. [Thanks to Frank B. for the Cisco press link.]
Open Park offers free Wi-Fi during inauguration: Park yourself on Open Park's network between 14th and 15th Sts. NW in Pershing Park so you can moblog, photoblog, and text-blog during the event. Unless the Secret Service decides that holding a camera phone up makes you a national security threat. This location is along the parade route.
JiWire tops 50,000 hotspots in its worldwide directory: I file this one under Self-Promotion because JiWire is my editorial and advertising partner. Although WNN is run independently, we work together on several fronts, and I'm just as excited as they are that their worldwide directory of hotspots has exceeded 50,000--more than 55,000 in actuality. London, surprisingly, has the largest number of hotspots of any city in the world, which is partly explained by the city's large metropolitan area under the same name, and partly by the incredible overlap of providers: BT OpenZone, The Cloud, Surf and Sip, T-Mobile UK, and others.
When I first met the JiWire folks in July 2003, I was impressed that they had brought together proximity, mapping, and hotspots. I like to be able to type in a street address, Zip code, or city, and then find out what's there, rather than reviewing lists. Lists become unwieldly, even though they're easier to scan, as the hotspot market grows.
Sascha Meinrath of CUWiN and Esme Vos of Muniwireless.com offer complementary insights on Indiana's frightening telecom bill: The bill in Indiana has elements that are familiar to me in my reading of Pennsylvania's and Wisconsin's legislation. The short story is that Indiana's law has the least protections for municipalities trying to offer wireless, telecom, and cable services. In fact, just the mere intention by a commercial operator to offer service within nine months--no commitment needed, nor do they have to build out in that period--scotches any municipality's plans.
Sascha Meinrath provides a detailed analysis of the bill's provisions noting how high the bar is set for municipalities and how easy it will be for incumbents to prevent municipal networks even in cases where one might imagine they'd be allowed. Sascha cites a case in Kentucky (via a colleague of his) that is precisely the kind of delay he's talking about.
Esme Vos looks at job creation and how municipal networks help communities keep or bring in new jobs. Telecom and broadband go hand in hand with the needs of companies. [links via GigaOm who also provides a link to the laws in various states prohibiting municipal broadband and related services.]
The Economist writes a call to open access to Broadcom and Atheros's radio technology for greater innovation (subscription required): The writer argues that by keeping their lower-level radio functions and any access to it close to their vest, they're discouraging wider uses of their chips and suppresses interesting projects from CUWiN and community wireless networking groups.
While the two companies produce Wi-Fi chips that don't use formally use SDR, they have aspects of SDR that make their concerns about opening up full control reasonable. And The Economist only suggests that more access than zero would be worthwhile. There is the Madwifi project which involves one programmer who was given access to the RF innards to write an intermediary, proprietary bridge between open-source drivers and the Atheros chips. But that's a pretty limited exposure.
Linux developers ask me all the time: when will Broadcom provide even that support? Perhaps The Economist's prod will cause both companies to think about how to sell more chips without incurring the FCC's wrath.
Mark Rakes notes that there's already an active thread discussing the article over the madwifi newsgroup.
Update: I want to clarify previous remarks a bit. From more technically minded types, I'm reminded to mention that the SDR that Broadcom and Atheros use doesn't allow access to all frequencies, as true SDR has the potential to do. Rather, it's SDR in the sense that there are several frequencies ranges, including both licensed and unlicensed, in certain chipsets.
Atheros and Broadcom should try to strike a balance in offering an abstraction layer which provides mediation so that open-source work could be built on top of it that still conforms to Part 15 rules but has a greater degree of flexibility than the current Madwifi project--and would allow any Linux use for Broadcom chips.
Another update: Sascha sent the link for the paper on which parts of the argument in the Economist argument are based, which he and two colleagues co-authored and delivered at a conference in Sept. 2004. I disagree with their argument that FCC sanctions a strawman; they can't be privy (nor can I) to the non-public aspects of working with the FCC and the issues surrounding partial SDR that might be part of the backstory to this issue.
Corriente offers Wi-Fi authentication at a small-office price: Corriente introduces Elektron this week, a $299 software package for Windows and Mac OS X that provides a full 802.1X and RADIUS account management and authentication suite using WPA for encryption and PEAP or EAP-TTLS to secure the login process.
The software is remarkably simple to use and can either pick up a list of users from the local machine on which it's running (Windows 2000, Server 2003, or XP, or Mac OS X 10.2.8 or later), or you can enter accounts directly. The company may add support later, possibly in a differently priced version, for external account support through a database or external RADIUS server. A fully functioning 30-day free trial is available at their downloads page.
Corriente's product is a tremendous price breakthrough compared to similar offerings by enterprise-focused companies that charge thousands of dollars and often per-seat fees for a server, or even Microsoft Windows Server 2003 which includes everything Elektron offers and a full server suite at hefty per-seat licenses coupled with complexity.
Elektron solves the out-of-band trust problem for using PEAP and EAP-TTLS by allowing you to use a certificate authority for Corriente on Elektron and client machines. A certificate for the Elektron server is then created against that authority. Elektron can export a Windows and Mac installer program for the root certificate authority.
This allows you to install a CA in all the clients that will connect to Elektron, which lets you avoid turning off a verification option in Windows XP that enormously reduces security. On the Mac, it avoids a step in which a certificate has to be accepted, although that's less onerous, and a user can confirm the fingerprint of the certificate against details provided by the Elektron server.
Elektron also allows external certificates to be used in a variety of ways. All certificate options provide for trust out of the Wi-Fi stream, which is key to any system in which security is your paramount concern.
The company estimates that it's practical to run as many as thousands of 802.1X clients against a single copy of their server on a modern (not server-grade) computer. What's not practical is managing accounts at that scale, which is why I say currently Elektron is best suited for smaller installations. A related issue is that best practices would require having two 802.1X servers up and running so that if one failed, the second was a fallback. A master/slave arrangement for synchronizing accounts across two servers would be a great addition in a future release.
Elektron becomes the third member of a troika of 802.1X options for smaller firms that lack the dollars, staff, or interest in using tools with more options or that are integrated as part of larger server packages. The other two cadres are InterLink's LucidLink and Wireless Security Corporation's WSC Guard. Both LucidLink and WSC Guard require special clients, but this allows them to control trust and fallback.
WSC Guard is a hosted 802.1X service which offers both WEP and WPA encryption, so it supports older client adapters without WPA upgrades or support. It includes software to run on the local LAN that the client communicates with in the event of an Internet connection failure. This allows a fallback from 802.1X to plain network encryption until the Internet link is back up. Without an Internet feed, Wi-Fi users would be unable to connect to the network. WSC Guard supports Windows XP and 2000, and they distribute a free generic WPA client for Windows 2000 that's made a lot of that platform user's in non-802.1X environments quite happy. WSC Guard is a subscription service.
LucidLink runs on a server on the local network and comes in home and small- to medium-size flavors that have fees based on the number of simultaneous users needing support. It, too, has a client, which runs only under Windows XP or 2000. Any user can attempt to connect to the server when he or she has the client installed, the server controls access including an out-of-band confirmation option. LucidLink requires WPA support.
Elektron, WSC Guard, and LucidLink each have distinct advantages, and the three together mean that smaller enterprises no longer need to question whether to use 802.1X--at least on some segments--but which option to choose. With the growing availability of VLAN setup in lower-end access points, 2005 should be the year of transition from WPA Personal-protected networks to WPA Enterprise security.
MIMO is all bursting out all over, so apologies for the headline: At CES, video was on everyone's mind hence the explosion of multiple-in, multiple-out antenna technology in which more antennas result in much greater bandwidth and enormously higher throughput. Belkin was first, and now Linksys (already), NetGear (soon), and D-Link (near future). PC World has a great run down.
NetGear will use an additional technology in some products called BeamFlex which incorporates elements of beam forming such as is used in phase-array antennas. Given how, uh, successful that's been in the enterprise market, let's see how it plays out in a home device. It's a neat idea, though, and can be used to assist plain 802.11g and MIMO.
Jeff Pulver, VoIP evangelist, conference planner, and gadget guy, says Las Vegas's free Wi-Fi filters: Ports blocked, me buckos, when you're stuck trying to leave CES. You can do Web, but you can't do telnet, instant messaging, or voice over IP, Pulver writes. There's ugly, ugly weather in Las Vegas right now, and one colleague reports that there's a three-hour line at the airport just to check in.
Actiontec introduces a two-USB and one-parallel port print server with 802.11g: They must not be certifying this particular device as they don't call it Wi-Fi. For $149 you can share with Windows computers (or Macs and Linux machines that support Windows-style printing) two printers connected via USB and one connected via a parallel port. This is a nice option for a home or office with multiple printers, and it reduces yet more cabling and complexity for managing printers on a network. It also means you can turn a computer off that might otherwise just be sharing printers.
Broadcom and Atheros's new easy-to-use Wi-Fi security enablers aren't as far apart as I thought: I was looking through the details this morning for SecureEasySetup and JumpStart for Wireless systems, respectively by Broadcom and Atheros, and found that while they work somewhat differently, they're closer than I thought in nature and intent.
Both systems try to remove the complexity from turning on encryption on networks. Broadcom's is more generally aimed at consumer electronics and devices with no real interface; Atheros's feels more computer- and adapter-oriented, but they make a good case about how it could be integrated into CE, as well. I spoke to Broadcom on Tuesday; Atheros this morning. Let's compare and contrast.
Initial setup. Broadcom has you push a button on the access point which causes it to create a key and store it internally, ready for the first client connection. This can be a hardware button on the device itself, or it can be a software button in a Web or client interface. Atheros has you connect to the access point via Ethernet to click a button, and create a password that's used to sign a key exchange that happens later. JumpStart also stores a WPA key.
Comparison: Neither version requires configuration of security options or key creation. Broadcom can push their button advantage, but consumers do need to run a wizard or other configuration software in any case to set up their DSL, cable modem, or LAN settings. By adding a step, Atheros increases security in the next step, but requires typing or key entry. Atheros also requires an Ethernet configuration (only for this stage) but derives additional security from this requirement. Broadcom can optionally run SES over Ethernet, but each device would have to connect via Ethernet. Both change the SSID and create a sufficiently long WPA key. Atheros's version can create an WPA2 (AES) key if that's available.
Conclusion: No advantage to either party at this stage in the process, although Atheros users might run into a problem if and only if they didn't have an Ethernet-equipped computer to handle the first step.
Note: It's clear that if ISPs worked with manufacturers, they could create a stub boot mode on gateways that would install the configuration for that client and enable the Wi-Fi security removing all LAN/WAN and security setup. This is the idea behind Microsoft's Wireless Provisioning System for hotspots.
Add a client computer. Broadcom has you push a button on the access point to put it into the right mode to communicate with a client PC Card. You can also use software on the access point to initiate this mode. On the client, users either push a button or use a client manager or menu to initiate communication. The AP and the client talk, the AP passes a key over a tunnel, and the client is ready to go. Atheros requires that you connect to the AP and have it start its communication mode. If the AP detects another JumpStart session, it backs off. Visual confirmation of the mode comes through flashing LEDs on the AP. On the client machine, you enter the same password used on the AP. This password is used to sign a Diffie-Hellman key that's used to establish a secure session over which the WPA key is sent.
Comparison: Broadcom leaves a window open here for an insertion in which a waiting client could grab a key from an AP before the intended client gets the key. There's no out-of-band confirmation that allows a rogue client to be rejected. Atheros, by using a password, increases complexity, but provides a way of securing the initiation of the SSL tunnel that's used to provide the key.
Conclusion: Broadcom lacks a Wi-Fi-based out-of-band confirmation option that would prevent malicious attacks from automated software that would attempt a denial of service on a user's network. Such software could be written because Broadcom and Atheros plan open standards. Broadcom does allow Ethernet to be used as a physically isolated and secure method of running SES, and it does notify users if a rogue client slipped in before the desired client connected. But there's no solution for a malicious DoS coupled with no Ethernet or no ease of using Ethernet. Atheros scores in edge cases with malicious DoS attacks.
Add a consumer electronic device, like a Wi-Fi DVD player: In Broadcom's case, push a button on the AP and push a button or trigger a menu on the DVD player. Atheros would require some kind of menu that would allow the entry of a key.
Comparison: Atheros seems to be at a disadvantage for entering alphanumerics on home entertainment devices without keyboards.
Conclusion: Broadcom may have a less secure method, but it does have a simpler process that will make CE adoption much smoother. On the other hand, CE devices may only have Wi-Fi and no Ethernet, which could make them more susceptible to being unable to join a network experiencing automated DoS.
In talking with Atheros this morning, they didn't convince me that JumpStart had a more secure end-to-end process. I'd already realized this and have a query out to Broadcom about the details. In home networks, it may be less critical that someone is ready to jump on, but an automated malicious attack is a real possibility for an open standard.
[Read the rest of this story...]
We Mac OS X users rejoice: Boingo has released their client for Mac OS X. I'm a Mac user--as well as a Linux and Windows XP user--and point out that the installed usage base of Wi-Fi on the Mac platform is ridiculously high among mobile travelers, thus making it a good way for Boingo to pick up customers. Boingo has the added advantage of having a Mac-friendly CEO, Sky Dayton, who founded EarthLink, bringing Internet access first to Mac users way way back in the mists of time more than a decade ago.
The client comes with a special coupon offer for a month's free service when you sign up for the unlimited $21.95 per month package. The client is required to negotiate the complexities of authenticating to dozens of different hotspot networks in the absence of a clearinghouse, which allows cell networks to handle cross-network roaming and settlement.
Today's podcast is a 6-minute interview with the editor of Mobile Pipeline, who is at CES: David Haskin, editor in chief of a fine publication for which I now write, was gracious enough to spare some time from overcoming his jet lag in Las Vegas to talk about what he's seen so far at the Consumer Electronics Show. We talk about MIMO and the proliferation of it, Wi-Fi phones, and UWB.
Some folks have reported difficulty in retrieving the MP3 file of the podcast if they don't have software that does it automatically, so I've provided tow links: plain MP3 file and a ZIP archive of the MP3. Both files are 1.3 MB.
Colleges are the bleeding edge of WLAN technology: I wrote this piece for Mobile Pipeline (my first for them) which started out as a more general best practices article for giant WLANs, and morphed into an article about college WLANs. Why? As I explain in the story, giant WLAN enterprises don't want to talk on or off the record about what they're doing. Academic institutions are happy to oblige.
Schools also have the disadvantage of having to support many, many generations of equipment and operating systems. You can't tell a professor to take a flying leap, and some schools may still have five-year-old laptops and desktops that still have to be integrated.
It's clear that 802.1X is the next big trend for both schools and enterprises. It solves many problems while adding tremendous policy flexibility. Using 802.1X and L2TP means that you can have smaller VLANs that are segregated by policy, group, or even randomly to balance users among VLANs.
DoCoMo is moving on up to Super 3G, then 4G: DoCoMo seems to have responded to the 3GPP discussions about 100 Mbps cellular networks by 2009 by talking further about its own plans. The company currently has a 3G network running W-CDMA with about 384 Kbps download and 128 Kbps upload. They plan to add HSDPA by March 2006 for 1 to 2 Mbps upload and download.
Then on to Super 3G in the 2007 to 2010 framework which they anticipate providing 100 Mbps down and 50 Mbps of raw speed, which probably translates to something like 15 to 20 Mbps down and 5 to 10 Mbps up. Quite impressive, but they'll be competing against fixed WiMax and 802.11n Wi-Fi clouds by that point. DoCoMo says Super 3G is mostly a software upgrade, which makes it a more feasible migration path.
Finally, starting in 2010, DoCoMo will roll out 4G with 1 Gbps downlink speeds.
It all sounds very pretty on paper, but DoCoMo has managed to execute in the past on high-speed cellular technology that has sounded slightly ludicrously ahead of its time until they launched it.
Did Kodak just build 802.1X into a camera? Yes! Kodak will release a camera in June that can upload photos via T-Mobile hotspots and any Wi-Fi network. Actually, the software to enable encrypted Wi-Fi connections, including the T-Mobile connection, isn't due until fall, for some reason. The new Easyshare-One sounds like a combination of Apple iPod Photo, PDA functionality (for wireless and previewing), and digital camera.
I had guessed that this camera's fall software release would leverage the 802.1X authentication that T-Mobile has added to its North American venues. 802.1X is both simple and hard. If Kodak preloads unique accounts, or allows people to set this up through PC or camera back software, there's very little complexity. The 802.1X supplicant in the camera can manage the connection.
A T-Mobile spokesperson confirmed Thursday morning that Kodak is, in fact, using 802.1X and that cameras would be preconfigured to work with T-Mobile's service through a yet to be determined trial period. Ultimate pricing is to be determined as well. The camera itself costs $600 plus $100 for the Wi-Fi card. This is the first consumer device that I know of that has plans to integrate 802.1X, and it could start a trend to add 802.1X authentication to hotspots and portable electronics. WiPod.X, anyone?
A very cool and strange feature is that you'll be able to browse galleries over the Internet using the card from Kodak EasyShare Gallery, the new horrible marketing name for what has been known as Ofoto for years.
It's a direct shot across the bow at cellular operators who are offering poor upload speeds on their high-speed network. Given that T-Mobile has articulated a long delay in their 3G rollout plans and don't want to clog their GPRS networks, this seems like a perfect symbiosis for Kodak and T-Mobile.
Look backwards, look forward: Marconi's first wireless test resembled UWB more than the frequency-huddled technology that was commercialized. Tom Standage, an editor at The Economist, specializes in this kind of Janus writing, in which the developments of the present are recontextualized in terms of their origins (sometimes lost) in the past. Likewise, he rehabilitates the past, by showing that historical technological innovation isn't all people huddled over a corpse, prodding organs.
Standage is one of my favorite writers and has a book coming out this year that will expand on this theme from the standpoint of beverages instead of technology with A History of the World in Six Glasses.
Roku has announced a version of its SoundBridge technology intended to be embedded in consumer electronics: Roku has received rave reviews for its stylish and technically adept--though expensive--devices for bridging sound and other media over wired and wireless networks. Their SoundBridge module includes all of the digital media playback technology and the pieces necessary for 10/100 Mbps wired and secure Wi-Fi networking.
There's a lengthy list of protocols, music formats, and DRM-based streaming music services they support: Microsoft Windows Media Connect and Windows Media DRM 10 and OpenTalk, UPnP AV, Rhapsody, Internet radio...WMA, AAC, MP3, AIFF, WAV and LPCM file formats...protected WMA content from music services like Real Networks' Rhapsody, Napster, MSN Music, Musicmatch and Walmart.com.
The only one missing? Apple's FairPlay protected version of AAC sold by the iTunes Music Store.
In related news, Roku has released a beta 2.0 update for its PhotoBridge HD1000 digital media player that allows it to play unencrypted AAC files.
On the heels of Broadcom's SecureEasySetup announcement, Atheros offers JumpStart for Wireless: Atheros trumpets three mouse clicks to set up a secure network, but they may have been trumped by Broadcom's announcement, which incorporates Linksys, HP, and Buffalo, and requires a single button to be pressed on an access point and client device, consumer or otherwise. The Broadcom button can be physical or in software.
Atheros describes JumpStart has having a four-layer security process, but it's not quite that fascinating: it requires out-of-band inspection (seeing LEDs flash), a tunnel that uses essentially SSL/TLS, a secure WPA or WPA2 (TKIP or AES) key, and a password for adding new devices. Broadcom's one-button method uses flashing lights, an SSL tunnel, and WPA.
Both Atheros and Broadcom's offerings work with Centrino--in fact, it sounds like both security simplifiers work with any client that can run the correct management software. The key part of the process lives in the firmware of an access point. Both offerings also work with Microsoft's new Connect Now profile management system, which is expected to be discussed in depth at CES this week.
Atheros also announced immediate availability of a MIMO chipset, adding yet another player to the MIMO market. By my count, there are now four, none of which seem to have any early interest in compatibility for future roadmapping towards 802.11n upgrades, and we're just getting started. I'd like to see a single statement from any of these vendors about whether their devices have any possibility of being firmware upgraded for 802.11n: my money is that none of them will take an upgrade to the final version and the no company will state firmly either way whether they'll be upgradable to final N.
I've been saying it for months and months: It's been crystal clear to me that Cisco did not have an internal WLAN switch strategy, and has its biggest problems in dealing with issues that switches can solve, which is policy-based VLAN assignment for WLAN users across network segments. News.com reports that a deal may be in the work for Cisco to buy Airespace, which is the leading marketshare vendor among the startup switch makers with seven percent of the market. I've thought Airespace was a 100-percent Cisco target, and am just surprised its taken this long to hear about a deal firming up.
Cisco's intelligence is in the access point, which means that hand-offs are coordinated at the AP level, making VLAN roaming and other related issues pushed out to the edge or handled by Ethernet switches, which doesn't work very well for mobile users. Airespace handles the logical part of this in the switch: the AP is a radio with some intelligence, but it's not the smartest part of the network.
Cisco has obviated some of its shortcomings in this area--and, of course, it has massive strengths in other areas--by turning WLSE, its centralized management tool, into as much of a switch-like controller for signal strength and other factors as it can.
Update: Om Malik has more commentary on this issue, noting that Airespace may rack up (pun intended) a large deal to upgrade Microsoft's network, and that Juniper might also be interested in acquiring Airespace. It's like the late 1990s, except with actual customers and revenue.
last night: A very fine article in Thursday's section deals with the increasing need for electrical power for the doodads that power our virtual life. The reporter recounts a personal incident in the third person about a Harlem diner owner castigating him for using their power--you can tell by the third person that it's the reporter; it's a very New York Times thing.
I once camped out in a hidden seating area of an Amtrak train from Seattle to Eugene to find a power outlet. I was trying to find a restroom, and noticed an entirely empty coach class section downstairs. I moved in and plugged in. At some point, an Amtrak employee noticed me and said, oh, this is for handicapped passengers. I looked around at the empty seats and said, I didn't realize. Do you mind if I stay until the seats are needed or should I just leave? She looked around, too, and said, well, no, I guess it doesn't matter. I was unmolested--except for a strolling Amtrak-sponsored troubadour, I kid you not, for the rest of the trip. [link via TechDirt]
Broadcom announced a one-button WPA security system for home networks and consumer electronics: The new revision of Broadcom's system to make it easier to secure home networks without entering settings is called SecureEasySetup (formerly SecureEZSetup), and is backed initially by Linksys and HP in a press release; Buffalo Technologies separately confirmed that they will also support the system through firmware upgrades.
SecureEasySetup in its simplest form requires manufacturers of Wi-Fi equipment to put an external button on their devices. Pressing such a button on an access point or gateway and then, within a short interval, on a piece of equipment like a Wi-Fi PC Card, a Wi-Fi-equipped DVD/media player, or Voice over IP (VoIP) Wi-Fi phone causes a secure connection to be created between the device and the Wi-Fi router. A WPA Personal key is then provided to the device, enabling it to get on the network with no manual configuration.
This one-button approach can be simulated through client software as well for older devices or devices that don't have the form factor for an external button.
The initial pass on this system last year was welcome but a bit wonky and didn't catch on due to too many manual parts of the process. Broadcom took a page from Buffalo's AirStation One-touch Secure System (AOSS) and took it down to the fewest necessary steps--and no typing.
"All you need to do is press a button on a router, then press a button on the client and then you're done," said Jeff Abramowitz, the senior director of marketing for Broadcom's home and wireless networking business unit. This new system is designed to help people connect all kinds of devices, he said, not just wireless LAN equipment.
He noted that while today most home use involves computers and a gateway, manufacturers are rapidly releasing devices that rely on Wi-Fi as a means, such as Wi-Fi in televisions sets, automotive entertainment systems, and VoIP phones. There's no easy way to enable security with many of these categories of devices. Printers, particularly, are a problem to bring onto a secured network as more and more come with Wi-Fi built in or available through a dock or adapter. Web-based configuration tools for printers and even access points are beyond the level at which most consumers want to interact with their equipment.
Abramowitz said that although their announcement today at the Consumer Electronics Show was made in conjunction with Linksys and HP, Broadcom expects announcements throughout 2005 from consumer electronics makers, VoIP handset developers, and even PC manufacturers who can add a SecureEasySetup button to laptops and desktops. (The button itself, by the way, has no functionality other than to trigger a specific piece of software rather than firmware.)
Read the rest of the story...
Belkin has many dance partners these days: Tom's Networking reports that Belkin has chosen Atheros for its A+G products, which may be heating up in the marketplace these days. Belkin uses Broadcom for 802.11g and Airgo for its "pre-N" MIMO devices.
Focus is showing its multiple-HDTV streaming technology that interoperates with the MBOA flavor of UWB: The chipset goal is 880 Mbps at 8 meters and 37 Mbps at 40 meters. These speeds allow many simultaneous streams over short distances. The company expects to sample their UWB chipset in the first half of the year with reference modules later in the year.
Three related articles indicate how fast 3G networks might become over time: Mobile Pipeline reported a couple of weeks ago on the latest HSDPA (High Speed Downlink Packet Access) network test. HSDPA offers a raw speed of 14 Mbps, which should offer perhaps 1 to 2 Mbps as a real speed for end users.
Cingular and Lucent also completed a UMTS/HSDPA test, showing that the current generation of equipment could be as fast as 3.6 Mbps with 400 to 700 Kbps typical for end users. Cingular is deploying a UMTS/HSDPA network with UMTS already available in several cities through its AT&T Wireless merger. The full rollout is apparently planned for 2006.
The Wall Street Journal reports that several firms, including Cingular, Vodafone, Motorola, Qualcomm, and Lucent are looking into 100 Mbps 3G networks possibly by 2009. It's interesting to see both CDMA and GSM firms meeting under the auspices of the 3GPP standards group talk about that kind of future speed. It means that it's clearly feasible. Of course, by 2009, we could have gigabit metropolitan ubiquitous WiMax or Wi-Fi, too.
Currently, UMTS offers a typical speed of perhaps the mid-300 Kbps. Verizon Wireless has committed to national coverage by EVDO, which has a theoretical top rate of 2.4 Mbps which translates to 300 to 500 Kbps in typical use.
In all of these cases, average speed and typical speed are two separate concepts. A typical speed is what a user might see most of the time, with large peaks (and occasional troughs) that skyrocket download speeds. Average speeds are therefore often higher than typical speeds if you were continuously downloading.
Upload speeds are also not consistently reported in any of these or similar articles. With EVDO, subscribers are currently limited to something in the range of 50 to 100 Kbps. This is fine for text and smaller images, but not for higher-quality phone cameras, video (even with high MPEG4 compression), or large file transfer, like PowerPoint presentations. [Thanks to Frank Bulk for a variety of links and insight!]
MIMO is cropping up all over: NetGear is the latest to announce a MIMO (multiple-in, multiple-out) product. They're partnering with Video54. Linksys has apparently chosen Airgo. Samsung and Athena announced single-chip silicon yesterday.
The near-term problem with MIMO isn't with the technology itself, but rather the non-standard higher speed modes. No one is claiming interference, but these modes won't be certified by The Wi-Fi Alliance or any other group. That means interoperability, even among devices made by manufacturers using the same silicon isn't assured.
The worst part, in my estimation, is that this generation of MIMO technology isn't what the final 802.11n specification will look like. So purchasing MIMO today except for specific applications in which you need substantially higher throughput and range from a single device over interoperable, certified 802.11g means you're buying a dead-end device.
802.11n might achieve speeds of several hundred megabytes per second, and there's is practically a total certainty that the final 802.11n won't be just a firmware upgrade, but rather a difference in both hardware and silicon compared to today's MIMO devices.
Linksys adds Airgo's MIMO technology in their WRT54GX: The new product isn't advertised as pre-802.11n, as Belkin's version is, but it does claim it can achieve as much as three times the range and eight times the throughput. Belkin's package of Airgo technology has gotten fairly positive reviews from many sources that like the performance that the router provides for non-MIMO cards. The one major exception is from David Haskin, editor-in-chief of Mobile Pipeline, who found erratic performance, especially in using the PC Card in non-pre-N environments. [link via Reuters]
In other MIMO news, Samsung and Athena Semiconductor said that they have an 802.11a/g compatible MIMO single chipset product. They expect speeds of up to 200 Mbps with their proprietary flavor that they'll be demonstrating at CES this week.
Engadget slips the news that Vonage ships its F-1000 Wi-Fi handset: The handset will work over hotspot networks allowing Vonage customers to use their service while roaming. Boingo and Vonage had a deal in place to test out VoIP over Wi-Fi hotspots, but it's unclear here in the late evening how that ties together. Also, Vonage's site doesn't yet list the announcement, so we don't know if they've partnered with various networks to ease authentication.
Authentication, or providing credentials that let you use a given hotspot network--whether a paid login or a WPA encryption key or handling 802.1X, even--is the biggest stopping block in allowing VoIP over hotspot to work. A technology like EAP-SIM, which would use a GSM SIM module to authenticate, might be one method of logging in. But it requires every hotspot or hotspot network that wants to allow this sort of connection to build the back-end to handle it.
There's slightly more coverage at USA Today which provides more background and notes the phone will probably be sold for about $100, but doesn't answer other questions. The article does make the point almost indirectly that the Wi-Fi phone could be a replacement for a phone-line adapter that Vonage now offers.
Wayport has 6,300 locations equipped; 5 million sessions in 2004: Wayport announced that they have exceeded 6,300 hotspots between locations with which they have directly contracted and those that they operate as a managed service, primarily for SBC Communications's FreedomLink network. The company also said that they had 5 million connections in 2004, with 600,000 in October alone. Wayport had about 1,000 location at the end of 2003, and a significant subset included in-room wired broadband hotels with Wi-Fi just in lobbies. Virtually all new locations are Wi-Fi only in retail establishments.
Dan Lowden, vice president of marketing at Wayport, said in an interview on Monday that Wayport believes they are now the biggest Wi-Fi network in the U.S., having exceeded T-Mobile's count, which is just under 5,300 according to T-Mobile's location finder. (This doesn't include T-Mobile's fee-based roaming partners.) "We're very very excited about it; it's a big milestone for us. We're excited to be the biggest in the United States," Lowden said.
Lowden said that Wayport is adding about 150 locations per week. They expect to exceed 6,000 McDonald's restaurants by third quarter of 2005, pushing their total over 10,000 locations.
Because certain large footprints have already been tied up, such as many coffee chains, and because of Wayport's exclusive deal for fast-food restaurants with McDonald's, Wayport is pushing into different kinds of spaces and services. For instance, their deal with Hertz gives them an alternate presence at airports. And Wayport will work with venues that may not need public Wi-Fi at all, or it's an adjunct to the real purpose: private services like cashless transactions (which they're doing with McDonald's) and security.
Lowden also noted that even with the hotel market so widely built out with high-speed services, there's still a lot of room to grow or rebuild. "We continue to see opportunities where folks have had three or four service providers over the last three or four years," he said. Wayport's now an established firm with major partners making it easy for them to bat clean up in cases such as these.
The company has grown fairly rapidly in the last year to handle the new contracts with over 330 employees now, with 150 hired in the last year.
We're trying to get serious about podcasting here at Wi-Fi Networking News: After some false starts in testing recording directly from Skype (a voice over IP software client), I believe we have a system that will allow us to offer regular podcasts--audio downloads via RSS--much more easily.
If you're trying to find a tool that will allow you to download podcasts automatically or view feeds with postcasts, visit Podcasting News.
The first Wi-Fi launch for Cingular is at the Raleigh-Durham (North Carolina) International Airport: This installation began as an AT&T Wireless rollout, and is now folded under the Cingular brand post-merger. Because Cingular is majority owned by SBC Communications, I expect to see both a quick incorporation of the AT&T Wireless locations under FreedomLink, SBC's Wi-Fi network, and a drop in prices.
Sure enough, Cingular has already lowered the egregious $70 per month fee for unlimited access to AT&T Wireless's tiny Wi-Fi network (mostly built with roaming partners) to $50 per month. Still outlandish, but less so--$50 per month is so 2002.