The writer says that air-to-ground service is like Wi-Fi in the sky, but it's using cellular data standards, and so it's much more like mobile broadband in the sky. He also writes that there's 3 Mbps, which is the combined up-and-down estimated throughput of AirCell, the only firm that can operate such service in the U.S. for commercial flights. The next graf mentions that satellite-based Internet access is coupled with, uh, 802.11b (yes, B) access points. I think that's an error, innit?

And the analysis of JetBlue's move is incorrect. The purchase of Verizon's Airfone network is about positioning equipment, not using out-of-date gear that can't be employed for phone calls on commercial airliners.

I'd suggest a more appropriate metaphor be used than the one in this sentence: "[Lufthansa] hopes the experience is more fruitful than its ill-fated 2004 deal with Boeing's Connexion service, which crashed and burned when Boeing shut it down two years later." Beyond the distasteful reference, Connexion was shut down in an orderly fashion, and Lufthansa was one carrier that loved it, and tried to get it to stay in operation, and, failing that, to build a consortium to revive it.

The article finishes with a set of incorrect conclusions:

"There hasn't been much news about how airlines plan to charge for these services." In fact, we know pretty much that it will cost roughly $6 an hour, $10 for a 3-hour flight or less, and $13 for a flight longer than 3 hours. That's from Aircell in various statements, and it appears to be roughly the charges expected from its competitors in the US. In Europe, mobile calls and texting prices are also known: about US$2.50 per minute for calls, and something like 25 to 50 cents for text messages, not much more than the egregious ground pricing.

"If the industry's cash crunch gets much worse, in-flight broadband might be mothballed before it even gets off the ground." It's unclear what part of the expense the airlines are bearing. In my discussions with firms over the last five years, it's clear to me that this round involves the providers bearing more of the cost--and hence the lower installation cost involved--but also retaining more of the revenue.

Wi-Fi a-go-go onboard buses: The New York Daily News checks in on the trend to put Internet access via Wi-Fi on board East Coast buses. The article notes that Greyhound's new sidewalk-pickup BoltBus service among corridor cities has provoked the long-running Chinatown buses to bolt on Wi-Fi as well. The Chinatown Bus Association says here that their bus tickets are cheaper and thus more competitive--but one of their members has already added Wi-Fi, and others are considering it. MegaBus also serves the coast and has Internet access, as well as DC2NY. The biggest problem, though? Passengers demand AC outlets, and only BoltBus has them on every bus. LimoLiner (New York to Boston) isn't mentioned here, but is one of the earliest firms I'm aware of with on-board Internet, starting in 2004, and they also have power to every seat.

Ozmo apparently is trying to leverage the ubiquity of Wi-Fi, the market reach of Intel (which has invested in the firm and is pushing its technology), and the dissatisfaction with Bluetooth device association and throughput to stick a wedge into Bluetooth's market domination. Well over a billion Bluetooth chipsets have shipped--CSR alone has shipped over a billion--and estimates put half a billion this year into cell phones alone. So there's a large embedded market to overcome.

This new technology, so far unnamed but apparently part of Intel's Cliffside research program, is trying to reduce complexity by reducing the number of standards needed to drive a computer, while increasing the flexibility of those standards. Ozmo and Intel's system would, for instance, allow a simultaneous WLAN connection and a PAN network of up to 8 devices using a single radio on a computer.

The press releases and articles make it quite unclear whether a new Wi-Fi chip would be needed; that chip would almost certainly not conform to today's Wi-Fi standards except in a compatibility mode, given that Wi-Fi has no capacity for PAN-style connections. Ad hoc mode isn't quite the same thing. In the past, extensions to the 802.11 standards that are the basis of the Wi-Fi certification and service mark were allowed as long as basic 802.11 worked as expected.

Bluetooth and Wi-Fi have been complementary technologies for several years. There were early conflicts--I wrote an article about the severe problems in using Bluetooth 1.1 and 802.11b back in 2001! But those interference and coordination issues were resolved, and Blueooth and Wi-Fi marched forward hand in hand, without any close association between the two trade groups behind the standards and branding, but with a lot of technology acquisitions and mergers on the part of companies that make Wi-Fi gear.

The Bluetooth SIG has been working for years to put Bluetooth on top of ultrawideband (UWB), which is still not readily available in the marketplace. UWB is always next year's big technology, and may be passed by except for applications like high-definition video streaming among a/v electronics. The SIG also announced support in Oct. 2007 for Bluetooth + 802.11, where a Bluetooth device could initiate high-speed transfers using 802.11 (yes, Wi-Fi, but not by that name; no partnership there). Bluetooth plus UWB is likely not available until 2009 at this point; BT and Wi-Fi, not until perhaps 2010. (See my article, "Bluetooth to Add Wi-Fi with UWB Delays in Mind," 2007-10-31.)

It's hard to see how Ozmo builds a place in this infrastructure, even with higher bandwidth, and what Ozmo says is lower power use and a lower cost for their chips, because laptop and desktop makers will need to buy into the Intel/Ozmo ecosystem. The demand for this kind of technology is typically driven by users who buy one component and need their computer to interface with it.

With Ozmo and Intel apparently planning to debut the Wi-Fi chips and driver support next year, it seems like a multi-year process to figure out whether Ozmo can evolve a competitive position to Bluetooth, even as Bluetooth is estimated to be embedded in over 1.2b cell phones by 2012.

Deciphering the Myths Around Complex Event Processing  by Ivy Schmerken stimulated a recent flurry of blog posts about the maturity of CEP, including; Mark Palmer’s CEP Myths: Mature or Not? and Opher Etzion’s On Maturity.

I agree with Ivy.  CEP is not yet a mature technology by any stretch of the imagination.  In fact, I agree with all three of Ivy’s main points about CEP.

In 1998 David C. Luckham and Brian Frasca published a paper, Complex Event Processing in Distributed Systems on a new technology called complex event processing, or CEP (Postscript Version).  In that seminal paper on CEP, the authors said, precisely:

“Complex event processing is a new technology for extracting information from message-based systems.”

Ten years later there are niche players, mostly self-proclaimed CEP vendors,  whom do very little in the way of extracting critical, undiscovered, information from message-based, or event-based, systems.  

A handful of these niche players have informally redefined CEP as “performing low latency calculations across streaming market data.”  The calculations they perform are still relatively straight forward and they focus on how to promote white-box algo trading with commercial-off-the-shelf (COTS) software.  In this domain, we might be better off not using the term CEP at all, as this appears to be simply a type of new-fangled COTS algo trading engine.

The real domain of CEP, we thought, was in detecting complex events, sometime referred to as situations, from your digital event-driven infrastructure - the “event soup” for a lack of a better term.    In this domain, CEP, as COTS software, is still relatively immature and the current self-styled COTS CEP software on the market today is not yet tooled to perform complex situational analysis.

This perspective naturally leads to more energy flowing in-and-around the blogosphere, as folks “dumb down” CEP to be redefined as it benefits their marketing strategy, causing more confusion with customers who want CEP capabilties that have zero to do with low latency, high throughput algo trading, streaming market data processing, which maybe we should call “Capital Market Event Stream Processing” or CESP - but wait we don’t really need more acronyms!

Hold on just a minute!  Wasn’t it just a short couple of years ago that folks were arguing that, in capital markets, it was really ESP, not CEP, remember?  Now folks are saying that it is really CEP and that CEP is mature?   

CEP is mature?  CEP is really not ESP?  CEP is really event-driven SOA?  CEP is really real-time BI?  CEP is really low latency, high throughput, white-box COTs algo trading?  CEP is really not a type of BPM?  CEP is not really for detecting complex events?   Complex does not really  mean complex? 

Come on guys, give us a break! 

(Anyway, no one is going to give us a break….  so stay tuned!)

Although several other quick-service restaurants like McDonald's lack any comprehensive Wi-Fi plan--Burger King, Wendy's, and Subway to name three of the largest--Wayport is locked out of working with direct competitors. This opens the potential for another firm to handle a several-thousand-location network. Wayport has worked with both McDonald's corporate-owned stores (about 2/3rds of stores in the U.S.), as well as reaching out to franchisees, who Lowden noted pay a predetermined flat rate for the service via McDonald's. "It's made them incredibly efficient to be able to offer this to their franchisees at one price, instead of variable pricing," he noted. Wayport acts as the layer between various telecom providers, applications and services, and the stores.

Wayport provides several kinds of back-office services, although credit-card processing was the first thing htey rolled out. They've extended to remote video feeds for security, Redbox DVD rental systems that are found in some McDonald's, and kiosks used for job applications. Lowden said Wayport offers things as straightforward but critical as a dial-up fail-safe when a broadband connection drops.

Wayport also manages AT&T's hotspot network, which puts them in the unwiring seat for the 7,000-odd Starbucks stores that will converted from T-Mobile to AT&T service during 2008. Wayport was once the clear leader in the hotspot builder market, with T-Mobile in the second position. Now, Wayport will be operating through a direct contract or management agreement over 18,000 hotspots in the U.S.; T-Mobile will likely be the second biggest with a couple thousand locations (Borders and FedEx/Kinko's tops among them). The No. 3 player is hard to figure. Panera?

I've been predicting for some time that media on the edge--music, videos, movies, and games stored on servers on the local Wi-Fi network--will be the next big development in venue-oriented Wi-Fi, with Starbucks likely far in the lead. Lowden wouldn't comment on any specific plans in the works, of course, but said generally, "Storing and caching all that content on the edge...hasn't been leveraged in the past, but it will be in the future to create a very unique experience." At Barnes & Noble, Wayport caches some multimedia data that's available to customers in the stores.

The advantage for in-store media storage is that you can leverage the speed of the local network, and add additional access points to distribute network load. The choke point is no longer the Internet connection, but local network speed. I expect--though Wayport, AT&T, and Starbucks haven't said it--that Starbucks infrastructure will be all 802.11n for this reason, likely with both 2.4 GHz and 5 GHz support for the best throughput in the higher-frequency band for media transactions. (In fact, I wouldn't be surprised if you could only buy movies via 5 GHz.)

Lowden also noted that the proliferation of mobile devices with Wi-Fi built in have led to them reaching out to venues that wouldn't have made sense for them to work with previously, and for unlikely candidates to reach out to them, too. Wayport is now working with a number of healthcare facilities that, while they have their own network infrastructure, wanted to outsource public access Wi-Fi (whether they choose to charge or underwrite it), and certain applications that they're not as experienced with running themselves.

A little history: In 2001 and again in 2004, the heat seemed to be on the public side of Wi-Fi: lots of money to be made, ostensibly, lots of partnerships and venues to be built, and an overcrowded supply of infrastructure builders. The year before, Wayport looked to be an also-ran in the hotspot provider business.

Despite being one of the earliest firms to put Ethernet and then Wi-Fi into hotels, and build out hotspots in airports; and despite their survival of the first hotspot meltdown in 2001 during the dotcom crash and brief venture capital shortage; and despite their early entrance into allowing wholesale pricing for hotspot aggregators; the firm seemed about to be eclipsed by apparently deep-pocketed Cometa (with AT&T, IBM, and Intel in various capital and support roles), Toshiba's mom-and-pop focused turnkey system, and T-Mobile, which had the Starbucks contract. What a difference a year makes.

Cometa, Toshiba, and Wayport contended for the contract to build out back-office and public-access service at McDonald's in the U.S., and Wayport won. Within a few weeks, Toshiba passed its few hundred locations to Cometa, which shut its doors in May 2004. Wayport, meanwhile, had cooked up a strategy for McDonald's that it announced later that month.

Their approach involved a fixed-rate charged for unlimited access by retail network partners for all the locations in their pool. This meant that partners had a fixed cost, instead of a per-session cost, and Wayport could obtain specific revenue even before usage by a partner ramped up. Wayport hasn't discussed the details of this arrangement in depth since, but has partnered with Sony with its Mylo, Nintendo with its DS game player, and ZipIt with its wireless messaging appliance.

The McDonald's deal also apparently gave Wayport a way to extend its work with SBC-later-AT&T; Wayport had earlier in 2004 became the managed-services contractor for SBC to build out The UPS Store/Mailboxes Etc. nationwide. (UPS dropped AT&T as its partner in mid-2007, although that didn't appear to have anything to do with Wayport's role.)

AT&T through Wayport developed its large resold/managed footprint that incorporated resale of Wayport's McDonald's locations with the UPS Store and a few hundred other managed locations, including a handful of airports. The Cingular acquisition of AT&T Wireless put more airports in SBC's hands, too. (SBC was once the 60 percent majority owner of Cingular; when SBC and BellSouth, the other owner, merged that put the newly rebranded AT&T in charge of Cingular which it relabeled as AT&T. Confusing, huh?)

Stalled-Fi in Florida: The Sun Sentinal newspaper looks at stalled, dropped efforts at city-wide Wi-Fi in Palm Beach County. Boynton Beach had a network early on, in 2005, but the city dropped the operator in March 31 due to complaints over maintenance. Delray Beach (E-Path) and West Palm Beach haven't advanced.

Minneapolis Wi-Fi requires booster for best use: This isn't an enormous surprise, or anything, and one of the consultants on the Minneapolis project said that USI Wireless starts with the notion that a booster is needed, which is highly sensible. Reporter Steve Alexander found service was highly variable outdoors with a standard laptop Wi-Fi adapter. The company sells boosters: a $160 high-gain laptop card and an $80 ($5/mo rental) home bridge. Alexander didn't re-test problem areas with the high-gain card. You can see the map of Alexander's test locations.

Orange Line in Los Angeles can't attract Wi-Fi operator: A spokesperson suggested riders should take advantage of "existing satellite" providers, where I think he'll be red-faced to know he should have said cellular. Or the reporter misheard. Say satellite and cellular each ten times fast. Now drink a glass of water.

Scarborough (Yorkshire Coast, UK) offers free Wi-Fi: 5.5m visitors pass through this coastal town each year, and a local business association has decided to unleash free Wi-Fi. The service will be pointed outwards for boats in the harbor, as well as inland.

Free Wi-Fi float in Sebastopol parade: The Apple Blossom Festival Parade last Saturday included "a fluorescent and sparkle-clad crew that shouted, 'Free Wi-Fi.' " The parade was led by a 1906 San Francisco Earthquake survivor.

The fiber types (such as multi-mode, single-mode), standards (SX, LX, LH) and  connectors (LC, ST, SC) seem to be a topics that need clarification about 80% of the time when we’re working with customers on networking equipment or site surveys.

Here’s a brief review of the various types of fiber, optics, connectors and when to use what. Let’s start with the basic stuff, and move down the line.

Multi-mode vs Single-mode
First of all, we have multi-mode and single-mode fiber. Multimode has a larger diameter ‘core’ or the area in the middle the light travels through. The larger diameter- think of it as a big tunnel- lets the light take different paths, creating multiple rays, or modes. The light bounces around more, which means the connectors and splices for multimode are more forgiving than for singlemode, but the bouncing causes dispersion and fidelity loss. On the other hand, singlemode has a much smaller diameter core, giving the light one straight path, or mode, through the cable. Because of this, singlemode offers higher throughput and longer distance, but the light equipment and connectors are much more finely-tuned. Which, of course, means singlemode is much more expensive.

When you’re adding or surveying multimode fiber, you should know what core size you’re working with. The core size affects bandwidth and the maximum distance you can reliably run it. Multimode usually comes in 50- or 62.5-micron, which is the core diameter. The larger the core size, the more bandwidth you get, but the shorter distance you’ can go. To give you a general comparison, most singlemode comes in 9-micron core, which is about 1/6^th the diameter of multimode.

When to use what. In short, the fiber type you choose will depend on 1) budget and 2) distance. Mostly, you’ll use multimode for short fiber runs, between switches, to servers and possibly between buildings, if they’re adjacent. You should use singlemode when you need higher throughput or a longer distance. Here’s a quick look at the types and maximum distances for each. I’ve also included a proprietary rating, for connectors using 1550nm wavelength over singlemode fiber, to get increased distance. (Standard for singlemode is 1310).

• Multimode - up to 220m with 62.5 micron core
• Multimode - up to 550m with 50 micron core
• Singlemode - up to 5km-10km (standard, using 1310nm optics)
• Singlemode - up to 70+km* (proprietary, using 1550 nm optics)
  

Fiber Optic Standards
You’ll need to know the type of optic to specify for your network equipment. Some vendors have their own proprietary fiber optics, but the standards are 1000Base-SX for multimode, and 1000Base-LX for singlemode. You can use multimode with 1000Base-LX with the addition of a mode-conditioning cable to set the light along the correct path down the cable. LX, which is standard, uses the ~1310nm wavelength. Vendors have created 1000BASE-ZX and 1000BASE-LH, which use the 1550nm optics to obtain longer distances. Note, here we’re talking about 1-Gig fiber, not 10GbE, hence the 1000Base. We usually just refer to these as SX, LX and LH, leaving off the 1000Base- when talking about the optics.

• 1000Base-SX - multimode
• 1000Base-LX - singlemode standard (can be used over MM with mode-conditioning cable)
• 1000Base-LH - singlemode non-standard (proprietary for longer distances at 1550nm)

Connectors
Here’s the fun part, and no one remembers what connectors they have (if they even knew in the first place!). There are several out there, but you’re probably going to only ever run into three - LC, ST and SC.

I’ll start with LC since that’s usually found on switches and other current network equipment these days. LC stands for ‘Lucent Connector’ (the creator) and is the connection type on SFPs (Small Factor Pluggable) or Mini-GBICs. They’re small, and were designed to replace the SC connectors.

Since I mentioned SC, let’s go there next. SC, or ‘Standard Connector’ are the predecessor to LC, and are similar in shape, but quite a bit larger. We suggest using the mnemonic ‘Square Connector’ to remember SC.

Last- and possibly least- we have ST, which really means ‘Straight Tip’, but many folks have a better time thinking of ‘Stab and Twist’. You stick it in and lock it in place by turning the outer barrel, sort of like BNC did. And yes, I’m old enough to remember the BNC days ;)

Duplex and Simplex
Most often, you’ll be using duplex fiber, which consists of a pair of fiber for bi-directional communication. Then- of course- you would use simplex fiber cables if you only need to send data a single direction. Those applications are more specific, but they do exist.

Ordering Fiber Cables
If we’re translating all our acronyms and numbers into something we can use, then let’s talk about how you put it all together when you’re procuring cables.

For example, let’s say you’re purchasing short fiber jumpers for connecting your patch cable to your switch. Most likely, you’ll want multimode, in a short length (2meters), with LC on the end going to the switch and let’s say SC on your patch panel. In our example, we’re assuming we have 62.5micron mm fiber.

What you’ll ask for is: Fiber jumper, 2 meters, duplex, 62.5-micron multimode, LC to SC.

LC	SC	ST

These are the best images I found to demonstrate the shapes and orientation of the various duplex fiber connectors we talked about. You can find these images and descriptions at Cables To Go.

Wowzers, I said this was going to be a short one. In fact, this post was originally titled “Fiber: A Very Brief Review of Cables & Connectors” but I had to rename it ;) Oh well- now you have all the information in one place for future reference.

# # #

I started Wi-Fi Networking News under the less euphonious name 802.11b Networking News back in April 2001 after spending months researching what became a front-cover article in Circuits, the then-separate tech section of The New York Times. The first post is still live, as are all the nearly 4,800 others.

(I had help: Nancy Gohring wrote part-time for WNN for a couple years when we had a bit more traffic; she took a full-time job for and still works for IDG News Service, which I am now slightly affiliated with through my new hardware regular blog at PC World.)

The site as it appeared in April 2001

Since starting, I've covered extensively the growth of the hotspot market, the rise and fall and rise again of municipal networks, the change in consumer equipment from expensive and slow to cheap and fast, the growth of the enterprise market, the phoenix-like in-flight calling/broadband market, and, more recently, cellular and WiMax technology.

Enterprise coverage was once a central part of Wi-Fi Networking News, but it became clear a few years ago that as equipment was redesigned to be integral to the enterprise, that my ability cover and test gear was too limited, and the need for true enterprise experience was necessary to write about it. This disappointed a lot of enterprise readers and equipment makers who wanted me to keep writing about corporate hardware.

The focus over the last few years on municipal Wi-Fi was not just necessary--few people besides me were covering it in depth--but also represented the only significant news in the Wi-Fi world outside of the development of 802.11n/Draft N gear. It's only recently that WiMax, cellular data, spectrum auctions, and in-flight broadband have picked back up to become stories that you all want to know about--because they've become real technology you might work with. As the city-wide Wi-Fi arc played itself out, I'm covering it less because there's less of interest; it's going to become routine and the province of city CTOs and CIOs.

While writing this site, I try to have opinions, but not an agenda. I try to keep an open mind, though I do descend into cynicism, often well founded, but perhaps too readily employed. I'll try my best to keep myself honest and cheery in the years to come.

The biggest trends I expect to see develop in 2008 to 2010 are in these key areas:

Appliances. I expected 2007 to be the year that Wi-Fi was in everything: cameras, games, phones, and tchotchkes. Instead, Wi-Fi has only gradually spread, with a few gaming consoles, and many handsets and smartphones gaining or extending their use. It may be that I missed a trend: cameras in phones may become so good by 2009, that we don't need a camera with Wi-Fi at all (Wired reports today on several 5 megapixel cameraphones shown at CTIA this week). It's also likely that if WiMax gets a foothold, we'll get handhelds probably in 2009 that sport high-speed connections for all kinds of high-bandwidth purposes, like live uploading of streaming video.

Video over wireless. I look at this category as not just another instance of broadcast, like Qualcomm's MediaFLO which is really TV to the cell phone; rather, we'll see ways in which Wi-Fi, WiMax, and cellular data are used to push stored and streaming media to all sorts of devices. I look to Starbucks, Apple, and AT&T to lead the way on cached media in stores that can be filled up at local network speeds: download a full-length, HD movie in a few minutes in a Starbucks from the iTunes cache rather than 3 hours at home.

Radio over Wi-Fi. Internet radio via Wi-Fi music players seems like a trend--buying a boombox you can tune in wherever you are, or using a handheld MP3 players--but even with many devices, I don't feel a sense that it's caught on quite yet. If Apple puts Internet radio over Wi-Fi into new iPhone/iPod touch firmware, it'll likely take off; Nokia allows a third-party program for its N series for Internet radio over Wi-Fi already.

Cellular data/mobile broadband. I admit to being wrong about the potential of cell data, due to the overhype from the carriers and the horrible pricing relative to throughput and availability of the 1xRTT and GPRS systems. As cell data networks have matured into true broadband--slow, but broadband--media, the hype has lessened, disclosure has improved (no more "unlimited" usage, eh?), and the value has increased. We'll see more of the same with faster flavors of GSM networking and WiMax's deployment. The networks will become faster and cheaper and less restrictive.

For a good sense of what people are still reading on Wi-Fi Networking News, here are the titles of the top 10 articles since I switched to Google Analytics in Sept. 2006:

• Change Your Linksys WRT54G Admin Password Right Now!
• WPA Cracking Proof of Concept Available
• Weakness in Passphrase Choice in WPA Interface
• Most Wireless Speakers Don't Live Up to Goal
• Best Wi-Fi Signal Finder Yet
• Linksys Latest Models: Your Experience?
• T-Mobile Loses Starbucks; AT&T Becomes Wi-Fi Hotspot Giant
• Editorial: Don't Buy Draft N
• WPA for Free under Windows 2000
• The L in Linksys WRT54GL Stands for Linux

A few observations. Security remains key in people's minds: Security articles from 2004 are still being heavily viewed in 2008. Linksys is definitely high in people's minds for particular problems: Change the default password, buy a Linux (not VxWorks) embedded router, report problems with various models. Oddly, the wireless speakers and wireless printers articles are short stubs that are pure blog: they link to longer articles elsewhere. The Best Wi-Fi Signal Finder Yet story is 4 years old and still gets 1,000 page views a month. The invisible hand--nay, the long tail!--works in archives as it does everywhere.

Will I still be pounding away 7 years from now on this site? That seems about as unlikely as the last 7 years, which means it will probably happen. Traffic has dropped off over the years from the time in which Wi-Fi was a great (and expensive) mystery to today when there's more information and less confusion about it. As long as there are any questions to be answered, I'll keep writing.

Wi-Fi over Ethernet combines electromagnetic resonance--the ability of a EMF to excite signals in wires--with excess wired capacity in a manner similar to how broadband over powerline works. Where properly equipped 802.11af Ethernet switches and adapters are available, coupled with WOE-capable Wi-Fi systems, the Wi-Fi signals will simply be picked up and carried by the Ethernet network. Switching and transmission then become limited to the extent of the wired network--which will improve throughput and range. (A future standard might allow passive powering of lightweight devices from Ethernet, which is a neat reversal.)

This is in the same category of new convergent standards such as Bluetooth over 802.11 and FireWire (IEEE 1394) over IEEE 741-2007: ways to provide better specs on one standard by combining it with another that has a complementary purpose.

Now, of course, modern computing systems tend to include gigabit Ethernet and Wi-Fi, so why do we need a third modality that combines the two? Partly because of new devices like the MacBook Air and smartphones like BlackBerrys with Wi-Fi built in. Without an Ethernet adapter, the range of these devices can be limited, and throughput restricted.

You were waiting for the magic number: How fast is WoE? Nearly 1600 Mbps raw speed, and about 30 Mbps of raw throughput. Before you scoff, remember that you might be able to use WoE over hundreds of meters across a switched Ethernet network, where a Wi-Fi signal might stretch just a hundred or two hundred feet. If Wi-Fi beats WoE, a computer will use Wi-F.

The Wi-Fi Alliance hasn't set the date of their certification yet, but I'm told it will happen any day. The mark will be added to the list of A, B, G, Draft N, WMM Power Save, and other symbols, as AF. The industry is considering a campaign around the phrase, "WoE is me(tm)!" trying to capture the excitement of the new synergy. Again, unfortunate acronym.

The IEEE has finalized and approved a draft, but final ratification isn't expected until 1 April 2009.

IronKey runs its own network of secure, anonymous servers that mask your identity. You can choose to change your exit point with a click, and keep track of throughput in case a given link is slowing you down. The IronKey plug-in for Firefox, invoked with a click, also stores all settings and caches on the flash drive.

Like Frye, I have long wanted to recommend an option for people who already use SSL/TLS protection for their email service, and don't need a VPN. IronKey appears to be the right recommendation.

IronKey works right now just with Windows XP and Vista, but their FAQ states they are are working on Mac and Linux components. IronKey comes in 1 GB, 2 GB, and 4GB configurations for $79, $109, and $149, respectively, including a year of "Internet protection," which covers secure browsing and a few other features. There's no information on the cost of the subscription fee after the first year, a notable omission.