In late 2004, the idea of city-wide Wi-Fi networks was electric. These metro-scale Wi-Fi networks would cross the chasm of the digital divide by bringing affordable broadband to low-income parts of major cities, and broadband of any kind to marginal neighborhoods, small towns, and largely rural counties. In mid-2008, the juice has drained out; yesterday, the last of the three major independent city-wide Wi-Fi network builders, MetroFi, said they're pulling the plug. EarthLink, another of the large providers, had already given notice of their exit in August 2007, and filed suit this week to remove the equipment for their flagship Philadelphia network. (Kite, a provider mostly in the Southwest, abandoned their Wi-Fi networks starting in early 2008.)

MetroFi predates the muni-Fi movement, having being founded in the early part of the century when broadband penetration via cable and DSL was still modest in many parts of the U.S., prices were high, and current and future speeds were low and expected low. Wi-Fi could compete admirably against these wired networks, it was thought, and against the weak first wave of third-generation (3G) cellular, on price, speed, and availability.

The incumbents don't stand still, and Wi-Fi, designed for interiors, didn't scale well. While it turns out to be possible to build a large-scale seamless Wi-Fi network that delivers from 1 to 4 Mbps of service outdoors to a laptop, and indoors through a $100 to $200 signal booster, it also proved true that a provider needed two to three times the number of Wi-Fi nodes across a city to achieve those speeds than was estimated when networks were largely bid out in 2005 and 2006. If you budgeted for 20 to 25 nodes per square mile and need nearly 50 of these multi-thousand-dollar transceivers, it's hard to imagine how that affects the bottom line. USI Wireless, which biult Minneapolis's network, appears to be the only firm that got the numbers and engineering to add up for them so far.

Artwork: Chip Taylor

I've written about two separate trends that have just collided today: in-flight broadband and fixed-rate unlimited monthly Wi-Fi roaming. Perhaps collided is the wrong word when referring to airplanes, though.

iPass, an aggregated remote access and end-point security provider, announced today that it will offer roaming with Aircell Gogo, likely to be the first in-flight broadband service launched in the U.S. At least two competitors are at work, but Aircell will likely be first with launches on American Airlines and Virgin America later this year, and maybe in a matter of weeks.

As I wrote recently, iPass provides service to corporations that can distribute their costs for metered service by the minute, hour, day, or month from all their employees across iPass's system that includes dial-up, Ethernet, 3G, and Wi-Fi worldwide. The company recently added individual packages that have fixed rates for U.S. or international usage. The cheapest plan is $29.95 per month for unlimited U.S. Wi-Fi, dial-up, and Ethernet (typically in hotel rooms) and $44.95 for the worldwide version.

iPass couldn't be pinned down today about pricing for the Gogo service, but expects to charge additional fees for Gogo access; a spokesperson said that prices haven't yet been set. Gogo plans to charge a retail price of $9.95 for flights of three hours or shorter, and $12.95 for all longer flights. (Correction: This article originally stated that iPass wasn't currently planning to charge extra for Gogo service, but that is incorrect. A spokesperson clarified earlier remarks to explain that additional fees will be likely, but that those haven't been determined yet.)

That's not out of line with the day rate at hotels and airports, where the walk-up rate can be from $7 to $15, but hotspot aggregators like Boingo (which owns many airport Wi-Fi operations) and iPass pay the provider a wholesale rate that can be as low as 50 cents per session. Wholesale providers and aggregators typically don't release these wholesale rate numbers.

Small businesses are treated like the last child picked for a baseball team in the broadband world. Too small to afford connections faster than DSL, which might be inadequate for their needs, and too large to find plans acceptable that, while affordable, don't provide them necessities like static, public IP addresses and business-level support. They're standing there with their ball and glove, waiting to play.

The AT&T U-Verse for Business service should therefore be of great interest to firms that want the fastest service with the least fuss. Cable providers have been offering business cable for some time, typically at just a slightly higher cost than residential service, but AT&T is pairing free hotspot office with very high-speed download rates, which could hit their offer out of the park.

The company is offering tiers from 1.5 Mbps downstream and 1.0 Mbps upstream for $40 per month up to 10 Mbps/1.5 Mbps for $100 per month. This includes providing a single Wi-Fi gateway for the office, and unlimited use on the AT&T Wi-Fi Basic network, which is Starbucks, McDonald's, Barnes & Noble, and airports that AT&T operates, but excludes hotels and roaming airports.

The service, initially offered in 40 markets served by AT&T, is enormously cheaper than what a T-1 line typically costs in the same place: usually hundreds of dollars per month for 1.5 Mbps/1.5 Mbps. T-1 lines can come with extremely high-level service level agreements, like 99.999 percent uptime, something that's not mentioned in today's announcement for U-Verse for Business.

That's part and parcel of this offering, though, which is focused on business users, not business servers. There's no mention on monthly data transfer limits, whether servers are allowed, or other similar issues. But for offices that need more oomphf and less expense, this should lower your bills, either through a switch to AT&T's service, or through increased competition from currently higher priced offerings.

Hate paying for Wi-Fi but want a reliable, decent connection that your corner coffeeshop might not be able to offer? (Mine can, but that's not uniformly the case.) Two developments this week may keep your pocketbook full. AT&T slipped out and Cablevision announced significant additions in Wi-Fi access for their current customers.

AT&T is taking over the Wi-Fi service from T-Mobile for Starbucks corporate-owned, standalone stores--over 7,000 in the U.S.--and slipped their kimono last week by accidentally (perhaps) making an iPhone-tailored gateway page available at Starbucks that prompted subscribers for their cell number. Enter it, and you were in, gratis.

That portal disappeared after a few days, but AT&T revised its iPhone plan features sometime in the last day or two to include access to all 17,000 of its domestic hotspots at no additional cost to iPhone subscribers. (That's 17,000 once the Starbucks transition is done, but T-Mobile and AT&T are engaged in a very goodsportsman-like handover in which subscribers to both networks will have access throughout; T-Mobile HotSpot subscribers will continue to have service for at least five years at Starbucks locations, too.)

AT&T already offers free Wi-Fi on its domestic network (excluding hotels and some airports, but including McDonald's, Barnes and Noble, and Starbucks) to its 1.5 Mbps and faster DSL customers, all fiber subscribers, and remote business access users--12 million in all!

(Just as I was about to post this, Computerworld's Gregg Keizer posted this story that AT&T had scrubbed that information from their site. Up, down, up, down, let's not call the whole thing off. It'll be back--but maybe not until the June 9 Apple developer's conference kickoff at which the iPhone 2.0 software, production software developer's kit, and 3G iPhone are all expected to ship or be released.)

Cablevision meanwhile dropped a bombshell today when they announced that they'd be building a $350 million--yes, million--Wi-Fi network across a big hunk of their coverage territory in New York, especially focused on Long Island. This service will be built over two years and be free to its millions of cable broadband subscribers, who already get among the highest speeds of any cable system in the US: 15/2 Mbps (downstream/upstream) and 30/5 Mbps flavors are their two listed offerings. Non-subscribers will pay to use the network, which they claim will have 1.5 Mbps of downstream Wi-Fi service.

Cablevision is building this network clearly to remain in play with a "quadruple" play: that is, voice, fixed broadband, video, and mobile broadband. Sprint and Clearwire's deal with Intel, Google, Comcast, Time Warner, and other cable operators has both a direct and indirect impact on Cablevision, which while not in competition with other cable providers, must fight back other video, data, mobile, and voice firms.

All I know is that additional services at no additional cost means a win for the consumer, and I'm happy to see it continue to expand.

 

Sprint and Clearwire have agreed to form a joint venture that will bring wired broadband speeds to mobile users starting as early as late 2008. The new venture combines all the spectrum licenses owned by the two companies into a new firm called Clearwire; Sprint will retain its cellular business and third-generation (3G) data network. The new venture is valued at an estimated $14.5 billion, which includes an infusion of over $3 billion in cash from Google, Intel, and several cable operators.

Both companies had separately charted paths to roll out mobile WiMax, a flavor of mobile broadband that's only been deployed in scattered places around the world, and is about two years behind earlier predictions for widespread deployment. The technology isn't monolithic like Wi-Fi, where all Wi-Fi devices are supposed to work together (more or less). Rather, there are many flavors of WiMax, including ones that can carry 10 to 20 Mbps downstream.

Sprint and Clearwire's WiMax rollout will likely offer speeds well above 5 Mbps; some reports put the peak rates at 8 to 15 Mbps, which is perhaps optimistic for an individual user, but might be reasonable for a pool available to all users in a given "cell" of coverage. Pricing isn't set, but Sprint expected to have production networks in place this summer, which was delayed to fall, ironically because they lack enough network backhaul to carry the high-bandwidth service from their cellular base station towers!

The joint venture and Sprint have both signed deals with all the various partners, which will allow Sprint to resell WiMax access, and the new Clearwire to resell 3G service on Sprint's existing network. Cable operators will be able to resell Sprint voice, 3G, and WiMax, which allows them to have a quintuple play (voice, data, and video over cable, plus mobile voice, and mobile data). Intel and Google have an option to resell service if they choose.

We'll start seeing WiMax adapters everywhere by 2009. Intel has been a backer of WiMax since they decided that existing cellular 3G plans wouldn't accomodate their needs as early as 2004. They will build Centrino designs that have WiMax and Wi-Fi built in. Companies like Motorola, Samsung, Zyxel, and others are committed to building residential and business fixed receivers, and laptop and desktop adapters, like PC Cards and ExpressCards.

The new operation estimates that 120 to 140 million people will be able to get their service by 2010, at which time AT&T and Verizon should start rolling out their 4G flavor, LTE (Long Term Evolution), which should offer comparable speeds. This increased competition among mobile broadband coupled with true broadband speeds that compare to today's wired services should force real price competition and improved customer service.�

Some laptop adapters will support both Sprint's current 3G service and Clearwire's 4G WiMax, allowing the fastest speeds as the WiMax network is built. Without dual cards and the cross-selling deals between Sprint and Clearwire, WiMax would likely never have had a chance in the U.S.

Fujitsu Siemens Computers will release LCD monitors this summer that use no power during standby mode, and charge the same price for their existing monitors. Standby mode typically consumes very little power with LCD monitors; an EPA Energy Star qualified monitor must use 2 watts or less in standby and 1 watt or less in off mode. Two watts can add up, however: that's nearly 9 kilowatt hours (kWh) per year, which can cost as much as one or two bucks, depending on your electricity prices.

You may ask: why does a computer pull power in standby or off mode, anyway? Isn't "off" off? Of course not. Just as in Douglas Adams's Hitchhiker's Guide to the Galaxy, where the eponymous guide's power-down feature changed at one point to "Mode Execute Ready," so, too, does all our electronic equipment that doesn't have a satisfying power switch clicker await our instant bidding. In off mode, monitors anticipate that you'll press the soft switch--a software controlled button--to turn them on; in standby mode, they're listening for signals from the computer to spring to life.

Fujitsu Siemens says that their monitor uses a simpler circuit that detects a signal from a computer, and that's brought their standby usage down to zero; the circuit is activated by the signal.

Five to 10 dollars over the lifetime of an LCD (perhaps five years) isn't enough to spend more for the feature. But companies that can include a zero-watt option may be able to shift firms spending the same dollars on their products instead of competitors.

The folks at Green Plug have just announced the winners of a messy plug competition that has a nice message behind it. Green Plug sells embedded technology for improving the efficiency of power consumption by electronic devices. They license at no charge their connector-type and power-supply communications protocol to encourage its use. Rather than having several AC-to-DC converters--the ugly power bricks that litter the undersides of our desks--Green Plug offers hubs that can flexibly support DC power output for many devices, dramatically reducing power usage.

As Google founder Larry Page noted in a somewhat unexpected approach to his CES 2006 keynote address, the complexity of power supplies is more than ridiculous. Google has, since then, talked about the inefficiency of power cords that connect power supplies to AC outlets--they can shed double-digit percentages of power as heat--and the ridiculous number of different voltages required on a computer mother board and the consequent inefficiency of PC power supply.

Artwork: Chip Taylor

We in the U.S., and particularly me in Seattle, Wash., suffer from the heartbreak of slow-broadband-paralysis. A large percentage of U.S. residents can't obtain speeds that are typical in Japan, South Korea, and some countries in Europe. You can easily buy 100 Mbps connections in Japan at an affordable price, and 8 Mbps ADSL is quite typical and cheap in the UK - it's thrown in as an extra for satellite TV and mobile phone service, for crying out loud. (BT is about to offer a 24 Mbps DSL flavor, ADSL+, to about a million homes this year and 10 million next.)

But broadband speeds and availability are finally starting to accelerate. AT&T, Verizon, Comcast, and others are rolling out 10 to 50 Mbps service across the country using fiber to the node and faster cable standards. Coverage areas for those services are increasing, even while rates below 10 Mbps become more widely available and faster, too.

In my home city, a territory controlled by Qwest, we're finally seeing DSL rates that match or exceed cable service. Qwest has a 14-state region dominated by rural areas, and the company has been in turmoil for several years following allegations about stock sales and financial results.

Qwest is now rolling out in 23 of its markets 12 Mbps and 20 Mbps DSL. They're the last major ILEC (incumbent local exchange carrier, formerly known as Baby Bells before mergers) to create a real plan for true high-speed broadband. Qwest charges $105 per month for 20 Mpbs service and $52 per month for 12 Mbps. A bundling discount of $5 per month applies. Upstream speeds don't seem to be disclosed on Qwest's site or in reporting on the matter.

Qwest isn't offering IPTV--just plain, fast broadband. This puts them in less conflict with customers who want to choose their own content, too.

Of course, neither my work nor home address qualifies yet for this faster service, which uses fiber to the node (an interchange point) rather than fiber to the home, which costs vastly more. DSL is used over the short interchange-to-customer premises link.

Our other incumbent, Comcast, has deployed its first 50 Mbps downstream/5 Mbps upstream service in the Minneapolis/St. Paul area, charging $150 per month for residential service. They plan to cover 20 percent of their territory by the end of this year. Comcast also said they'll have 100 Mbps service within two years and 160 Mbps service after that point.

None of this puts us rapidly in the same category as admittedly countries with smaller area and even more population in the densest portions. But it's nice to finally see a bit of motion. The next step is dramatically lower prices. I'm not holding my breath until competition heats up, probably as wireless broadband starts to hit in 2009 and 2010, which could push costs down for sub-10 Mbps service, forcing wireline providers to increase speeds and lower prices.

Sharp will start selling the X-series of 37-, 42-, and 46-inch HDTV sets in Japan next month that include a paired set of wireless transmitters for high-definition video using technology from Amimon. Amimon uses unlicensed frequency in the 4.9 to 5.9 GHz range, varying by country, to pass uncompressed 1080i or 1080p signals (both video and audio) over as far as 30 meters or 100 feet indoors. These are the same ranges used by 802.11a and 802.11n Wi-Fi flavors worldwide. (A double-wide channel, or 40 MHz, is required for 1080p; 20 MHz for 1080i.)

Wireless high-def is currently in development in many forms, that includes using Draft N- Wi-Fi-like technology as Amimon is deploying; 60 GHz millimeter-wave micro-antenna arrays with SiBeam; and several firms working on incompatible overlays to the WiMedia ultrawideband (UWB) wireless technology.

Sharp's announcement marks the first time that an electronics company is offering a line of products with built-in hardware paired with an external transceiver, however. Most of the work to date has focused on cable-replacement paired transmitters that cost as much as $700 and eliminate the need for an HDMI cable. Sharp is offering the wireless transmitter that connects to the source as an optional accessory, but it's still a big step forward in reducing the clutter of home-entertainment cables.

Sprint just starting shipping an interesting new USB mobile broadband modem--the Sprint Wireless Compass 597. The modem, which accesses Sprint's cellular data network, is extremely small, has built-in GPS, contains a slot for a micro-SD (Secure Digital) storage card, includes all installation software, automatically activates itself, and works equally well with Mac OS X and Windows XP and Vista.

Sprint is pricing the modem aggressively, too: it's $50 when purchased through their online site with a two-year service commitment. Sprint charges $40 per month for a 40 MB per month limit ($1/MB additional) or $60 per month for unlimited use, which the company told me yesterday is truly unlimited. They only restrict abusive uses, such as operating a server or running peer-to-peer file transfers.

The adapter supports EVDO Rev. A, which is nearly universal across Sprint's EVDO network now, and is advertised as supporting averages rates of 600 Kbps to 1.4 Mbps downstream and 500 to 800 Kbps upstream. The modem also handles 1xRTT, which works closer to dial-up modem speeds, outside of the 16,000 cities (encompassing 246 million people) and 1,500 airports that have EVDO service.