Mesh Projects & Gear


Over at open-source mesh pioneer LocustWorld, I discovered that Ultramesh is helping to implement the mesh networking software being installed in NW Louisiana and Stevenson Washington. Here’s a post from Joe Schlick of the Stevenson Wi-Fi Project:

The Stevenson Wi-Fi Project was conceived as a way to promote tourism in the town of Stevenson, Washington, about 1300 people in the heart of the Columbia River Gorge in southwest Washington.

We were able to convince our city council to grant promotional funds to our Stevenson Business Association to provide free public access wi-fi internet connectivity throughout our downtown core and along the public areas of our Columbia River Waterfront (the total area covered by our mesh is approx 4 x 8 city blocks).

The idea is that this free amenity will encourage tourists to stay longer in our community since they are able to stay in contact with projects and people that might otherwise require them to move on. The second, and potentially greater, benefit is our ability to now promote ourselves as a forward thinking, technically savvy community that understands how business in the 21st century is, and will be, carried out. Hopefully attracting non-resource based, living wage paying jobs to an incredibly beautiful place only 45 minutes from a major metropolitan area (Portland, Or.).

Our project was funded in July 2003. We ordered equipment in August and had the basic mesh operational by October.

We have 6 nodes at this time with two more being installed by the end of January. One of these is located on the roof of our county courthouse and will have a webcam overlooking our downtown, waterfront, and across the river to the mountains in Oregon.

The mesh is designed so that all public gathering areas in our downtown core have connectivity. That includes our parks and restaurants. We are working closely with our city business association as well as the county chamber of commerce to incorporate the Stevenson Wi-Fi Project in all promotional efforts. One thing is to build it – the other is to get the word out and start benefiting from it.

I will try to keep you posted as the project expands. Anyone interested in further details can email me at jschlick@earthlink.net

Joe Schlick
Stevenson Wi-Fi Project

FastLine Internet may have been the nation’s first wireless mesh network using LocustWorld technology. They began offering broadband service in May, 2003. They’ve unwired the towns of Vivian, LA and Linden, Texas, spending about $25,000 servicing 1/3 of the town and can grow the mesh to cover the rest.

“I’ve been setting customers up with 3 levels of speed. I have a residential plan where they can purchase the wireless gear and pay $24.99/month. I have another plan where they can pay $39.99/month and we supply the gear. There is a plan for commercial accounts where they pay $59.99/month, we supply the gear. I charge $30 to install and configure the gear that we install. That is probably way too cheap, but we want to get everyone in town”.

Police use Vivato in Salida, Colorado. One antenna covers most of the downtown area. Hermiston’s secure wireless cloud, built by EZ Wireless, covers 600 square miles, encompassing four counties and seven cities for use by the Oregon Chemical Stockpile Emergency Preparedness Program (CSEPP). Another seven cities will be added this summer.

Medford uses MeshNetworks’ broadband solution. The largest city in Southern Oregon at 70,000, it has approximately 150 police officers and 75 fire personnel. Viasys Services (a MeshNetworks VAR) did the systems integration, deployment and project management. Garland, Texas has the largest mobile mesh network in the world. They also use MeshNetworks’ technology. Mesh networking has also been deployed in Baton Rouge and other communities in the South.

PacketHop, a spinoff of SRI, completed a multi-agency, interoperable communications exercise led by the Golden Gate Safety Network (GGSN) — a coalition of federal, state and local Bay Area public safety agencies. Under the leadership of the California Office of Emergency Services, PacketHop’s mesh networking enabled instantaneous mobile broadband connectivity, first-of-its-kind situational awareness applications, across diverse devices, difficult terrain and disjointed networks. But mesh gear is not standardized. Gear isn’t interoperable. An IEEE 802.11 Working Group met in Vancouver, British Columbia, this January to begin an industry-recognized standard for wireless mesh networking.

WiFi Plant says it is likely the IEEE will try to encompass both ad hoc/client mesh, where a signal can be extended over multiple hops, and full network/infrastructure mesh, where the self-healing/self-configuring connections are used for the backhaul that serves individual clients. Where the standard will truly go remains to be seen.

A mesh standard could be defined in 2005, but likely won’t be finished until 2007. Meanwhile, incompatibility between ALL mesh vendors is the rule.

AT&T Bailing from CoMeta?


WiFi Netnews says AT&T has sent email to customers that the Cometa relationship has ended.


“The email says that AT&T Wi-Fi Service will no longer be offered as of yesterday. It implies that it’s service is entirely driven by Cometa, but AT&T operates service through NetNearU and Concourse Communications at the Newark airport. Perhaps that’s in the past tense. AT&T’s Wi-Fi page is still showing locations, almost all McDonald’s in the New York tri-state area, and no Newark airport listing.

…The announcement says that new options for wireless Internet access will be available starting in March, including daily options, but it doesn’t define those as Wi-Fi. Will AT&T have no Wi-Fi in their future? The core AT&T doesn’t seem to have interests in data networking outside of the enterprise, that I’m aware of.

…This announcement may not signify anything of importance for Cometa, which has failed to produce an announcement of any scale since last fall. AT&T was the company that they pointed to as Cometa’s major national partner for reselling access (in New York at McDonald’s and in Seattle at 250 locations — although those Seattle locations aren’t listed at AT&T’s site), but AT&T was hardly a Wi-Fi reseller”.

It is big news (if it’s true). AT&T is a major investor in Cometa along with IBM and Intel.

I’ve always assumed that AT&T (the long distance business), needed a “last mile” line back into homes and business, especially since it lost the cable connection. WiMax (perhaps with VoIP), could have been AT&T’s vehicle.

But AT&T announced this week they will launch VoIP nationwide. Why would AT&T bail? More news could be forthcoming from CoMeta.

Free Geek


Here’s a story from the front page of the Portland newspaper about an innovative community organization recycling old computers.

You can get a computer for free (with instruction) by donating 24 hours of volunteer time to recycle old computers. It’s called Free Geek. Here’s part of the story [edited for brevity from The Oregonian].

Oso Martin launched Free Geek on Earth Day 2000 with a stack of misspelled brochures and a Web site describing an organization that was more virtual than real.

No nonprofit had done exactly what Martin envisioned: Collect old computers, teach volunteers to fix them, give the good ones away to good causes and make sure the rest are responsibly recycled.

There were times, especially in the first year, when Free Geek almost didn’t get off the ground. Finding people willing to donate old computers wasn’t the problem. Paying for a place to store and refurbish them was. At one point, Free Geek owed $12,000 in back rent. Still, the landlord and others were enchanted by the ponytailed Martin, who drives a beat-up truck fueled with 20 percent vegetable oil biodiesel and talks really fast when he’s excited.

Today, Free Geek not only is current on its rent, it has expanded its space twice. The organization occupies an old Langendorf Bakery building that stretches across a southeast Portland city block. Last year, when Oregon employers were either laying people off or locked in a hiring freeze, Free Geek boosted its staff from six to 13 plus added four paid internships.

The U.S. Environmental Protection Agency estimates there are 315 million obsolete computers, monitors and other peripherals in the country, containing an estimated 1 billion pounds of lead, 2 billion pounds of cadmium and 400,000 pounds of mercury. Most everyone agrees that these toxic materials should not be disposed in the country’s landfills, where they might leach into groundwater. There’s also agreement that it’s best not to ship these old machines to a developing country, where improper dismantling can threaten both workers and the environment.

There’s widespread disagreement, however, about how obsolete computers should be handled and who should pay for it.

In June 2001, the EPA convened a panel of experts representing government, the electronics industry and environmental groups to develop a national strategy for dealing with computer waste. The panel, known as the National Electronics Product Stewardship Initiative, held its final scheduled meeting earlier this month in Portland, ending three years of talks without developing a plan everyone could agree upon. Industry has committed to keep talking.

In the meantime, with no national solution to computer waste and old machines piling up, Oregon and other states likely will come up with their own recycling and disposal rules. Martin is a member of an Oregon state electronic waste advisory committee that has held its first meeting. He’s not waiting for government or the electronics industry to dictate Free Geek’s future, however.

At first, Martin looked for a job in architecture. The only thing he could find was designing medical office buildings for Kaiser. Been there. Done that. He finally settled on doing some consulting, and he whispers when he says, “I only had to work four or five hours a week.”

He earned enough money to cover rent and buy food for himself and his dog, Jake, a black and white Border collie.

Martin decided to commit the balance of his time to the social and environmental causes he believed in. For a while, he and Jake spent 50-plus hours a week volunteering for City Repair Project, a group of activists dedicated to creating public gathering places. Martin helped organize the 1998 “Hands Around Portland” event — which drew thousands of people for a symbolic protest against what activists then characterized as a “growing isolation brought on by freeways, gated communities and unregulated technology.”

One day, Martin and a fellow activist, Matthew Follett, sat looking at the half-dozen or so computers piled high with books and papers in Martin’s dining room.

“I really should get around to fixing the computers to give them away,” Martin said.

Follett’s response: “Why don’t you just do it, then?”

Today, Martin has 16,600 square feet of space in which to fix and store old computers and electronic equipment. And the machines keep coming.

Free Geek’s warehouse is filled with central processing units and computer monitors stacked from the floor almost to the ceiling. There’s a separate room for printers — known affectionately by the Free Geekers as “printerland.” There’s also “stereoland,” as well as a yet-to-be-named pile of laptops, phones and other gizmos taking up space Martin would like to see turned into a free Internet cafe. Someday, he’d also like to have a “museum of weird stuff.”

A private donor gave Martin enough money to start Free Geek. But the organization won its big break when it received a $40,000 grant from the state Department of Environmental Quality and the city of Portland in 2001. Other public and private grant money has trickled in.

Then, Free Geek received $159,000 in the summer from the Meyer Memorial Trust, a Portland-based foundation that supports a broad spectrum of nonprofit efforts in Oregon and Clark County, Wash. The money is to be used to expand Free Geek’s staff and operations and to help other nonprofits learn to use Open Source Software, freely available applications that are easily customized.

Lately, Martin has been asked to help establish a Free Geek operation in Lancaster County, Pa. Another guy e-mailed from South Bend, Ind., asking whether he could use the Free Geek name for his computer recycling operation.

Sure, Martin said. No charge.

“Oso has this rare vision, and Free Geek is a magical organization,” says Marie Deatherage, program and communications officer for the Meyer trust.

In fact, Deatherage says, if the state of Oregon is looking for people to feature in its “We love dreamers” campaign, state marketers ought to put Oso Martin in the ad.

“He’s the guy they’re talking about,” she says.

The Man Who Invented Television


“I know that God exists. I know that I have never invented anything. I have been a medium by which these things were given to the culture as fast as the culture could earn them. I give all the credit to God.”

–Philo T. Farnsworth

Who Invented Television? Sparks of Genius abound. Paul Nipkow’s mechanical television goes back to 1884.

restored sequence by D F McLean   1998 In 1925, after years of development, a Scottish inventor, John Baird made practical demonstrations in London. A 1933 disk may be the earliest known recording of a television show.

In the 20s-30s, Philo T. Farnsworth “captured light in a jar” and is widely considered the inventor of television. Vladimir Zworykin developed it in the early 30’s. By late 1939, sixteen companies were making or planning to make television sets in the US, but the new sets were often incompatible using a variety of scan lines. The BBC actually made the first “HDTV” broadcast in 1936 (using 405-lines).

Farnsworth came up with the basic technological concepts of electronic television at the age of 14, while preparing a potato field with a disc harrow, and transmitted the first television image in 1927 at the age of 21. He was subsequently embroiled in a long and acrimonious patent dispute with RCA, which he eventually won, after one of his old schoolteachers came forward with a diagram of a basic television system that Farnsworth had drawn for him as a teenager.

The widow of Philo T. Farnsworth, Elma “Pem” Farnsworth, celebrated her 96th birthday this week. Mrs. Farnsworth granted Interactive TV’s Tracy Swedlow an interview.


[itvt]: Your late husband was Philo T. Farnsworth, the inventor of television…

Farnsworth: To me, by the way, he was “Phil.” He took the “o” off his name, because in grade school they called him “Fido.” He was Phil long before I met him.

[itvt]: You were obviously closely involved in the birth of television, and you have witnessed some of the most important historical developments in the history of communications and media.

Farnsworth: Yes I have.

[itvt]: Among other things, you were the first person to appear on television.

Farnsworth: That is true.

[itvt]: How did that happen?

Farnsworth: It was in our laboratory in San Francisco. I guess the reason I was the first person to appear on television was because I was the only other person around in the lab at the time.

[itvt]: Did your husband realize how important his invention would prove to be?

Farnsworth: What he said is that the television would make it a smaller world, because one country could see how the other country lived, and therefore countries would be able to solve their problems around the conference table instead of on the battlefield.

[itvt]: Did he ever talk to you about how he saw television developing in the future?

Farnsworth: One thing I remember him saying was that eventually there will come a time when a television set could hang on the wall like a painting.

[itvt]: The laboratory you mentioned was part of a small company–quite similar to one of today’s high-tech start-ups–that was set up by your husband, yourself and your brother, correct?

Farnsworth: Yes. We had gotten $25,000 in funding, and Phil had promised to have a transmission within a year. He succeeded, because he thought it out so well. He saw the problems that would happen–and there did turn out to be quite a few problems.

[itvt]: What kinds of problems did you encounter?

Farnsworth: Well, of course, the main thing was to get as many lines per inch as he could, because the more lines per inch you had, the clearer the picture. We spent a whole week in the Los Angeles library trying to find out just how fast you would be able to send a picture to fool the eye into seeing it as a picture, rather than as a lot of dots. So he had that to start with.

[itvt]: What was the first image generated by these electrons–i.e. the first TV image?

Farnsworth: The first image was a straight, horizontal line.

[itvt]: How much time did it take to move from being able to show a horizontal line to being able to capture your face?

Farnsworth: That line was received on September 7th, 1927, and in 1928 he was showing a full picture. In 1929, they made a film transmitter, and were able to have a continuous transmission, so that films didn’t jump from one frame to the next.

[itvt]: I understand that one of your husband’s proudest moments was when he saw the TV images that were beamed back when the first men walked on the moon.

Farnsworth: Yes. Phil turned to me and said, “That has made it all worthwhile!”

[itvt]: I take it that your husband told you how he came up with the idea of television? I heard that he was riding on a tractor, and that the idea just came to him.

Farnsworth: Actually, he was driving a 2-horse team. He was in a field planting potatoes. But, as Phil would say, he never invented anything: he was the conduit through which these things were given to the people.

[itvt]: So he felt that his inventions just appeared in his mind and that he simply communicated them to the public?

Farnsworth: Yes.

[itvt]: He had little formal training in electronics, correct?

Farnsworth: He used to say that invention was as if people were building a temple and had had to put up elaborate scaffolding to build that temple. They would get so involved with the scaffolding that they lost sight of the temple. He hadn’t had much formal training, and so he didn’t have all these preconceptions that many other people had–preconceptions that made it hard for them to let go of their scaffolding and actually build the temple.

[itvt]: The story of how he had to defend his patents against RCA is well-known. Those must have been trying times for you.

Farnsworth: When Phil gave a talk, he had slides with diagrams. Well, there was always a front row of people with cameras, and RCA would take the photographs those people took, build the equipment he was showing, and then go into interference with him in Washington. Now this was a very expensive thing to go through, and they had wrecked a number of lives that way. They were trying it with him, but they finally had to give in and take a license.

RCA used to say, “We collect royalties, we don’t pay them.” He was the first one to make them take a license.

Localizing Consensus Plans


Local governments are facing a decision. They must replace the analog 800 Mhz CDPD networks used by mobile terminals for police and fire. Now they’ve got homeland security checks to do it.

What to do?

They’ll have to move to more expensive cellular-based GPRS or 1XRTT data networks and pay cellular carriers [perhaps $250K/year] for the privilege, not to mention the cost of new radios. Or they could invest that money and create their own do-it-yourself Wi-Fi network.

Many will opt to use the unlicensed 2.4 GHz band for police and fire communications. They might use mesh-based 2.4 GHz units like Tropos, wireless LAN switches like Bandspeed and Vivato, or use dozens (or hundreds) of APs combined with automatic handoff between GPRS and Wi-Fi. City vehicles might use wireless USB clients with 802.1x or WET-54 bridges with mag-mount antennas and 2.4 GHz amplifiers.

Is this a good idea? Should “public safety” depend on the 2.4 GHz unlicensed band? Should these networks support free “public access”?

It depends. Consider the impact of Nextel’s Consensus Plan. There may be plenty of new (licensed) frequencies available for public service. But it will take years for that to happen.

Local governments don’t have years. The CDPD/cellular networks, on which public safety workers depend, may be turned off this year or the next. Cellular carriers want to force the move to GPRS/EDGE. They expect that data revenues will grow enormously. Nevermind that cellular data will likely always be more expensive than “IP” data — taxpayers will foot the bill.

Some kind of “local consensus plan” might be formulated. A migration strategy. It would need to benefit everyone, public safety officials, city officials and the general public.

My plan:

  • Build a public safety 2.4 GHz network.
  • Provide a “free” component for shared public use.
  • Move to licensed frequencies when available

Here’s the thing; 802.16 is coming. Intel thinks WiMax could be used to deliver high-speed wireless Internet access throughout a small city for about $100,000, one-tenth the cost of rolling out fiber-optic lines today. WiMax has a theoretical range of 70km but in practice can deliver 10Mbps over a range of three miles. Schools and community centers could use it. It makes a better “city cloud”. Wi-Max is the future. Let’s not muck it up.

There may only be a 3-5 year window for effective use of a 2.4 GHz public safety/”freenet” before it gets overcrowded. Then public safety users can move to their new licensed bands and use long range 802.20 networking with 1Mbps connectivity. It will likely be cheaper and faster than cellular alternatives and they could control it. The 2.4 GHz network might be turned over to a coalition that runs the thing with “equal access” provisions. It could be upgraded to WiMax/802.16e on the city right-of-way.

OneCleveland, led by Case Western University, launched last October, now has over 1,400 access points. It will foster applications in five areas: bridging the digital divide, health care, arts and culture, scientific research and e-government. Hermosa Beach has a city-funded “cloud” that provides free wireless broadband to homes. The installation will cost the city between $75,000 and $85,000 to provide free access to every one of the city’s 18,000+ residents, with annual operating costs totalling around $18,000 per year. WirelessHoustonCounty will “unwire” a whole county.

Consensus anyone?

Freq Consensus?


Verizon Wireless objects to the Nextel Consensus plan that reallocates Nextel’s emergency-radio-band and is calling for regulators to auction the wireless spectrum instead. The FCC is hoping to adopt an order in its 800 megahertz band proceeding by the end of the first quarter “if possible,” said Wireless Telecommunications Bureau Chief John B. Muleta.

The rapid growth of Nextel’s cell-phone service has caused an increase in interference in the nation’s public safety radio frequency. That’s because the FCC gave Nextel the ability to do cellular communications on their bands which are interleaved with public safety radio used by police and fire. Where once Nextel towers were miles away, now they’re across the street from fire stations or blocking an emergency call in the field.

To eliminate this disruption of emergency communications, the FCC is considering a proposal from Nextel to vacate their adjacent 700mhz and 900mhz frequency in exchange for a swath of higher bandwidth at 1.9 gigahertz. Nextel calls it the Consensus Plan.

The proposal calls for Nextel to exchange 16 megahertz of spectrum spread around the 700 MHz, 800 MHz and 900 MHz bands for 6 megahertz in the upper 800 MHz band and 10 megahertz in the 1.9 GHz band. If approved, this would leave Nextel with 16 megahertz of contiguous spectrum in the upper 800 MHz band, on which the carrier could continue to offer its voice service, and 10 megahertz in the 1.9 GHz band to offer “4G” services in the future.

Nextel has offered to pay $850 million for relocation of their interfering operators in the public safety spectrum, which were installed in good faith under FCC approval.

Other than Verizon Wireless, Nextel has had little opposition to the plan. But critics and industry observers say it would be a great windfall for the No. 5 player, giving it a heaping supply of network capacity at a relatively low cost.

Verizon Wireless says the chunk of 1.9ghz spectrum that Nextel wants is worth $7.2 billion and should be auctioned rather than given away. For its part, Nextel says its so-called consensus plan is a fair trade because it would be giving up valuable spectrum in exchange for a replacement spectrum.

Verizon Wireless is urging the FCC to put the public radio property up for auction, but the company did not promise it would be a bidder. Verizon is more than $40 billion in debt and facing the potential need to buy out partner Vodafone’s 45% stake in the joint venture.

Verizon clearly doesn’t want to concede any ground to a competitor, and has managed to get 23 members of Congress to sign a letter opposing Nextel’s swap proposal.

Observers say the FCC wants to clean out the public safety frequency quickly and at no cost to taxpayers. While giving Nextel some prime wireless turf angers rivals, it represents no direct cost to taxpayers and it achieves the goal of a clear emergency radio band.

The FCC rep declined to make any predictions as to when the agency would decide, but said that early this year was a safe bet.

Nextel offered to give up 10.5 MHz of spectrum on the 800 MHz band for 10 MHz on the 1.9 GHz band, in the Consensus Plan. That plan would redesignate 1910-1915/1990-1995 MHz to Nextel.

MMDS (between 2.5 GHz and 2.7 GHz), has 200 MHz of licensed spectrum. It’s shared with Instructional Television Fixed Service (ITFS), a band of twenty (20) television channels available for educational institutions.

MDS at 2.1 GHz was originally used to distribute, via microwave, a single channel subscription television service, usually HBO, to home subscribers. Nextel is testing Flarion for “4G” service in the MDS band. That system, which is often said to be the basis of the evolving 802.20 standard, requires licensed frequencies, below 3.5 GHz. Data rates would be mobile, as with 802.16e, but range would likely be better using lower frequencies, in the 700-800 Mhz band.

Nextel has won wireless licenses, spectrum, tower leases and other related assets from bankrupt Nucentrix Broadband Networks for $51 million in an auction. Nextel also bought Worldcom’s MMDS licenses for $140 million. That portfolio gives Nextel about 1/3rd of the 2.5 – 2.6 GHz MMDS licenses in the United States – as big as Sprint’s MMDS portfolio.

DailyWireless has more on Nextel’s Consensus Plan, Wi-Max Outed, 4G War News, FCC Created Interference, 4G Clouds in the United States, 802.16e Vrs 802.20, Public Safety Communications and FCC’s battle of the bands.