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.
Meraki has changed its pricing and feature model for its mesh networking system, angering early users: Exiting its beta, Meraki has changed its pricing and service model, while requiring the display of advertising and a piece of the action for handling billing. This abrupt change, announced quietly last week, has resulted in a nascent networker revolt. It may be that early infrastructure builders abandon Meraki because to continue expanding networks, their cost structure has gone way up while control has gone way down.
The two basic nodes that Meraki sells, Indoor and Outdoor, were priced at $50 and $100 during the beta. The beta flag is gone from their site, and while you can still buy the nodes at that price, many features formerly included raise the price to $150 for Indoor and $200 for Outdoor. These prices move them into the range of what used to be much higher-priced metro-scale hardware.
Meraki's system works by having little intelligence built into nodes instead relying on an Internet-hosted administrative system to handle the heavy lifting. Adding nodes requires just some configuration via a Web site. Each node auto-discovers existing networks allowing their addition. Each gateway to the Internet added to a Meraki network increases the bandwidth pool.
The change puts tiers into effect. The Standard edition with the cheaper pricing doesn't allow billing nor user authentication, nor does it offer a custom splash page (a logo is all). Even the network name gets Meraki branding on it ("Free the Net" precedes a local name). Standard does offer a private WPA-protected network for the owners, but without any granular access control. This means that a group that wants to change small amounts for access or control access by user cannot use the entry-level version. According to the forums, this includes a lot of early networks. Standard networks also can't "whitelist" more than a handful of devices, which means that devices that lack the ability to display a Web page for authentication purposes can't even access the network. So no Xboxes, VoIP phones, and so on.
The Pro edition bypasses these limitations, essentially adding a $100 per node license for advanced control and the ability to bill. However, because several of the Pro features were previously in the only version Meraki sold during its beta, many networks had build their business model, their pitch to cities and communities and businesses, and their financing on $50 and $100 nodes. The Pro version ups the ante while taking away some of what these network builders need, according to the forums. (A Carrier class version, with no pricing disclosed, allows a tie-in with existing user authentication system, and allows more of a private-label version.)
So-called Legacy networks, ones that were built before this last week, can continue to run as they do or be upgraded to Pro status at no extra cost. But once upgraded, all the Pro limitations are in effect. Any new nodes added to a Legacy network must be Pro nodes, too; there's no option to add Standard nodes. This is part of what's causing ire on the Meraki forums. (One of these networks, Kokua Wireless, was praised in an editorial in today's Honolulu Star Tribune; someone from the network noted on Meraki's forums, "It'll be hard to build a business even @ the Pro level knowing Meraki can make such drastic changes." One network operator has already flashed his Merakis with alternate firmware.) Update: Meraki says it's been talking to all the operators and groups that have problems with their new model; I'll have more details Monday.
In this formal release, Meraki now displays advertising on every page using what they variously call Community Messaging, Community Messaging & Advertising Platform, and Messaging Platform. It's often euphemistic. In effect, Google ads--Google is a Meraki investor--will appear on every page in a special toolbar that requires no installation. Which means that it's inserted using some technique--probably frames--that might cause other problems, too. Network operators can insert localized messages, but to a limited extent. There's no option for Standard and Pro users to disable advertising or the messaging bar. There's no disclosure on what revenue, if any, Meraki splits with the network operator. If Meraki handles billing for Pro users, they keep 20 percent of gross revenue.
I had to check the Web site after being alerted to this change, assuming that senior management--largely a bunch of MIT grad students who turned RoofNet into this clever commercial offering--had been fired or shuffled, and a new carrier-oriented CEO was hired. As far as the site shows, it's the same gang. Which surprises me, given how open the group was about communicating, and how interested in community building. This is a huge stumble, when the community tells me that they were unaware of the changes to come until the Web site started to have new information on it.
At $150 and $200 for managed nodes, Meraki has now lost some of its edge against companies like Tropos Networks, the most likely comparison. While Tropos, Cisco, BelAir, Motorola, Strix, and SkyPilot typically charge $2,000 to $5,000 for their nodes, with the more expensive ones including 5 GHz backhaul radios, those prices are heavily discounted, several sources have told me recently. Tropos nodes can cost EarthLink, its biggest customer, under $1,000, I have it from multiple parties outside the service provider.
While $1,000 is clearly five times $200 for a similar outdoor piece of gear, that belies several factors. Tropos is using enormously high-powered radios, capable of producing up to the maximum legal signal strength in 2.4 GHz. Tropos has a sophisticated back-end that allows integration with many popular management and authentication systems. Using Tropos or other vendors' network gear is substantially less trivial than building one from Meraki pieces, but with enormously larger coverage areas relative to cost now, Meraki has moved itself towards that equipment range without bringing the robustness needed.
The system and hardware has gotten rave reviews for the simplicity in building a network in bits and pieces. But the good will may have just flown out the window. I have just sent a query off to Meraki's PR firm, but it being Sunday morning, I don't expect to update this post until Monday.
Meraki's recently announced outdoor repeater with optional solar/battery power could transform grassroots networks: Meraki has established itself as the cheapest provider of hardware and back-end controller support for mesh networking. With its $50 indoor nodes and new $100 outdoor repeater (shipping in July), the firm is making good use of its $5m in investment from Google and Sequoia Capital.
In an interview this week with co-founder and chief executive Sanjit Biswas, I discovered that despite the low cost, Meraki is "actually able to seel these devices at a small profit," Biswas said. The insides of a Meraki router are pure commodity, rather close to what's found in an inexpensive Linksys, he said. "We spent a lot of time studying the economies of scale opened up by them."
Biswas noted that he and his co-founders were graduate students for the five years prior to starting Meraki, and had to hack small, cheap gear by necessity. In the process, they put the software intelligence--the controlling part of the network--into remote servers, leaving individual devices to act as relatively low-powered nodes in a network that would self-form as it grew.
The new outdoor repeater is designed to allow those network to expand even further. With a built-in antenna that offers higher gain than their indoor node, they expect a range of several hundred feet. Optional higher-gain antennas can be installed, too. "Really what we're doing is saying, hey, we're a software company; if you want to tinker with the hardware and pick up some long-distance antennas, then we're all for it," Biswas said. Some customers have already set up 10 to 12 km links with the pre-release versions of the repeater.
The Meraki Outdoor has two 10/100 Mbps Ethernet ports to allow local networks to be tied in or to run back-to-back repeaters with highly directional antennas. For now, each device has a single radio, but Meraki--like virtually all wireless LAN and mesh/metro firms--uses Atheros chips, and Atheros recently announced their roadmap with a reference design for an access point with two separate radios.
Meraki's outdoor devices are highly ruggedized, Biswas noted. "We designed the outdoor to survive anything from a monsoon to a scandinavian winter--pretty robust conditions." Meraki's equipment is already used in 1,000 networks around the world, and there's already a need for the outdoor product to spread coverage from the inside out, as Biswas explained.
The firm developed its own custom routing protocol that builds on top of the 802.11 set of standards. This allows each node to receive and send normal Wi-Fi traffic while also transferring control information. Their mesh routing system optimizes to reduce interference rather than for the highest thorughput. "People in general using Internet access only use tens to hundreds of kilobits per second when you average it out," Biswas said in the kinds of installations their gear is used in. Groups of mesh nodes can switch channels to improve network quality or increase throughput, however.
The intelligence of the system lies in Meraki's servers--these servers provide controller features that can cost thousands of dollars with other mesh or large-scale networks on top of the higher cost of each node. Even Ruckus's recently announced small-to-medium-sized business networking equipment requires a controller that starts at $1,200 for a small number of access points. (Ruckus is focusing on throughput and signal efficiency, however, for ensuring the best quality of service with streaming, voice, and overall data transfer.)
While the mesh nodes exchange information about path efficiency, Meraki's servers handle auto-discovery, collect statistics, and provide remote management. This means that any cluster of Meraki nodes--any group of nodes on a unique channel--must have its own backhaul. But because the system is self reconfiguring, no network will accidentally sever itself by losing an Internet feed on a segment.
Each Meraki network becomes its own cloud of access with unique private addresses assigned persistently to devices that join the network. This allows someone to roam from node to node, and it also allows multiple injection points--nodes connected to an Internet feed--to allow aggregating of bandwidth. Each TCP/IP connection can be routed the most efficient way through the network. The system can also packet-shape and throttle, allowing each node to control how much of its bandwidth is used by the Meraki network. That's a critical factor when shared DSL or other connections are being tied into a network.
The outdoor repeater includes power over Ethernet (PoE) as a standard feature, including the necessary adapter. This can allow an Ethernet cable to be used to provide juice to an outdoor node that's not sited near an electrical outlet. Using electricity outdoors or on rooftops is always a tricky civic code issue, too, making PoE a better choice.
The solar charger/battery combination that will be offered for the outdoor repeater should allow use in places where electricity and Ethernet just isn't an option, whether on a condo rooftop, on a savannah, or in a park. Biswas said that they were fortunate to have on their team the right combination of engineers to slice the cost of the device down to roughly $400 from $1,000. The final price hasn't been set. "This is almost a hobby project we had running internally," Biswas said.
Solar panels are actually in short supply, driving up the cost. Through use of a custom charge controller built in house, Biswas said they cut the panel size to a third of their initial requirements. They also built Ethernet into the controller, allowing it to handle reporting of a range of statistics back to their servers. This will allow network operators to pull useful details about the panel and its battery without making a remote visit.
The batteries are lead-acid, which have a number of downside compared to lithium-ion, but which are relatively cheap and available worldwide in the right format to be replaced out on site without expensive international shipments of custom items. "We're able to use batteries that are used in security systems around the world," Biswas said.
With a battery-operated Meraki Outdoor node being recharged by solar panels, network builders of all stripes are freed from a grid requirement. Biswas said that their work on a growing free network in San Francisco led them to see this as a developed-world problem, not just a developing-world one. His customers "want us to deploy larger and larger networks; they were having trouble with the power because it's a sticking point. People would get hung up on, hey, I'm putting a power outlet on your roof," he said.
(The San Francisco network has seen over 3,000 users during its operation iwth about 250 to 300 users on during peak periods. It covers about a square mile, and it's let the company get their hands dirty, Biswas said.)
Their new outdoor product may be the biggest challenge yet to the more expensive--but more robust--nodes sold for thousands of dollars by metro-scale equipment makers like Tropos and SkyPilot. Biswas said that they weren't attempting to challenge these firms in the municipal market yet, as they have no public-safety offering, and don't view their equipment as resilient enough for that yet. Tropos, he noted, offers bullet-proof casings for their nodes. "They're hooking them up to cameras and people do shoot at these things."
For now, Meraki's early customers are entirely through word of mouth. The firm has no sales department, and continues to devote most of its efforts to extending its product line, and getting nodes into the field. I see remarkable potential for small networks to be built easily by neighborhood groups, community organizations, and business associations--and then later linked up with either more Meraki equipment or higher-end metro-scale gear.
Whatever the case, Meraki has broken the price barrier and dropped the complexity bar. Will a million neighborhood networks bloom?
New York City's parks will eventually be unwired: Dana Spiegel, executive director of veteran community wireless group nycwireless, spoke to me yesterday about the group's efforts to put Wi-Fi in New York parks, the challenges with that, and what's happening in Central Park. We also spoke about an RFP issued by the economic development arm of the city that will examine the state of broadband across all the boroughs and what might be done to improve access to the Internet to all residents. [40 min., 20 MB, MP3]
A Champaign-Urbana-based group gets $500K from the National Science Foundation: The Champaign-Urbana Community Wireless Network (CuWIN) group in conjunction with the University of Urbana-Champaign (UIUC) will use the grant to continue to develop their open-source mesh routing protocol system over the three-year life of the grant. The group wants to produce software that creates self-configuring, self-healing mesh networks with advanced properties that would allow clusters of nodes in which some cannot hear all others can still function with a high degree of reliability.
The New York Times reports that the vendor selected in Oct. 2004 to put Wi-Fi in several parks must compete the Central Park job by July, others by late summer: The company, Wi-Fi Salon, is quoted in the article as sounding less sure about the deadline than the city is. Wi-Fi Salon won a bid after Verizon backed out for Battery, Central, Flushing Meadows-Corona, Pelham Bay, Prospect, Riverside, Union Square, Van Cortlandt, and Washington Square Parks, and Orchard Beach in the Bronx, the article notes.
So far, Wi-Fi Salon has put service in place only in Battery Park. Their plan has to been to find sponsors to foot the bill and provide the $30,000 per year minimum required in the contract. Future contracts will require no revenue to be paid, apparently.
The article notes that NYC Wireless has put Wi-Fi in a number of parks over a number of years, and was selected to unwire the Dag Hammarskjold Plaza in conjunction with a non-profit associated with the plaza.
Let's recall that a year ago, Wi-Fi Salon's founder Marshall Brown told Wi-Fi Planet that his firm "will be launching eight access points (APs) in Central Park in July , all of which will offer free Internet access to the public using a backhaul of roughly 64Mbps..." This is the same interview in which he made the claim--which I hope he has since learned was understated by several orders of magnitude--"I'm saying to the city of New York, each of these APs can support 255 Wi-Fi hotspots each. If each Wi-Fi hotspot covers 300 feet in circumference, I could cover a square mile of open public location and I have 20 locations." [Photo via ladymay79]
Interesting, other coverage of this story seems to state that the city has just approved Wi-Fi for Central Park. For instance, the BBC reports that the city "announced plans." Their story (and many similar ones) make it sound like corporate sponsorship is already lined up. They say the Parks Departments "says that by July" service will be in place rather than the facts, which is that this is a vendor deadline.
The first summit hosted by CUWiN was in 2004, and apparently connected a lot of dots (and faces) for those focused on politics and others on technology: This year's event, March 31-April 2 in St. Charles, Missouri, splits tracks for technology, implementation, and policy. Based on the previous event, there's very little talking heads up on a stage, and more symposium-like interaction. The cost ranges from $65 to $675 depending on type of registrant.
In this New York Times piece, I look at the transformation of community wireless networking advocates from hardware hackers to political operatives: The first wave of community wireless networking (CWN) groups appeared around 1999 and 2000, and this first wave inspired a larger wave that followed. While CWN initially focused in many cities on installing hotspots and helping to set up free locations, the larger themes have taken over as hardware as gotten cheaper to buy and easier to run.
When I started thinking about writing this article months ago, I thought I would be writing an elegy for community wireless. It seemed to me that membership had dropped, groups had disbanded, and leaders had left their positions. Instead, after talking to a few dozen people, many involved since the early days, I discovered that the focus has shifted away from the brute force stage and into subtlety.
In the early days, most groups were talking about how to create antennas, build node maps, modify firmware, buy gear for cheap, and get locations hooked up. Some were thinking all along about building their own citywide networks; others just wanted to convince all manner of venues to offer service for free.
Over time, it's become so easy to create a Wi-Fi hotspots or even a zone spanning a fair amount of area, that the challenges have shifted to issues like network neutrality, or making sure that everyone can use a network without prejudice for purpose or equipment. Many of the ideas of community networking have found their way into municipal proposals, and many of the wireless advocates I spoke to have tried to shape these proposals--often successfully.
Seattle Wireless probably represents one of the highest achievements in the area of neutrality, because they've built a network of what looks like now over a dozen nodes that use an open-source mesh routing protocol to create a neutral medium. Anyone can plug in multiple times into any point on the network to create tunneled services across the entire redundant, optimum route system. There are no rules on what the network is used for, which makes it unique. Their new Capitol Hill location gives them one of the highest points in Seattle from which their antennas can be "seen" and thus employed. (A bad bit of phrasing I wrote in the article makes it sound like the tower is their only location; it's just a centrally located, very high one.)
Wibiki joins Fon in reiterating failed business models of the past: They suggest installing their client software on your computer and server software on a Linksys router that can be flashed with their firmware. Their approach is similar to Fon's--and the one that Sputnik, Joltage, and SOHOWireless tried years ago and left--in which there's a pool of users in an authentication database and each access point has its own separate backhaul paid for by the person running the server software.
The Wibiki folks (part of Speedus) suggest that by creating an environment, a business model will follow. They ask on their blog, "How can we create sustainable business models to make Wi-Fi ubiquitous and free for users?" [link via Yeah!Fi]
Brief report from Sascha Meinrath of CU Wireless: "As of a few minutes ago I got the go-ahead for deploying Community Wireless folks within the New Orleans area. We've secured a base of operations and are working with Part-15 to get FEMA approval to operate in the emergency area. We've got people heading down starting tomorrow, so if you are interested in being part of this team, drop me an e-mail. Prometheus radio has received dispensation to set up an emergency LPFM station in New Orleans, so we're interested in anyone who would like to help with that. Finally, in addition to people and equipment, we're also going to need donations to help support the on-the-ground Community Wireless team." Their Web site has an ongoing blog.
Donate to wireless team:
Part-15.org is mounting an enormous effort and is looking for donations and volunteers. Anyone in the wireless industry should contact them if you want to help.
Of course, the Red Cross and other organizations are raising money like mad to help feed, clothe, and temporarily house 100,000s of people--and later help in more permanent ways. My wife and I have contributed a (for us) significant amount to the Red Cross. But there's a lot to be said for giving money to help bring wireless service and FM radio to New Orleans to allow communications. I want to see people fed and safe first, but part of bringing order is bringing contact.
In related news, most of the cellular operators report a significant restoration of their networks in and around the disaster areas as they've been able to restart generators and repair equipment.
The Champaign-Urbana Community Wireless Network (CUWiN) released the fruit of their efforts at the end of the week: The project is an open-source effort to provide mesh networking with no center. The system is self configuring among nodes which need no non-volatile or permanent storage.
To set up a CUWiN network, you burn a CD with the 0.5.5 software later this week and use it to boot a computer with a supported wireless card. The system finds nearby nodes, creates tables, and establishes itself as part of the network. The software is free and open source.
The full press release is after the jump.
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.
Mike Outmesguine is organizing Post-Tsunami Reconnect: The effort will try to get wireless equipment and expertise to areas of need in the tsunami disaster. Information can be a powerful tool in organizing resources in the midst of chaos. Without a telecom or data infrastructure, the distribution of food, clean water, and medical help could be directed to places that don't need them. Best of luck on this effort.
Direct donations of aid, by the way, are being collected by various groups, including Oxfam and the Red Cross. Amazon.com has devoted its home page to the Red Cross's efforts, and has helped them raise over $1.2 million so far. (The International Committee of the Red Cross's site is down at the moment, apparently overwhelmed by traffic.)
Technology Opportunities Program (TOP) granted money to a number of community wireless groups: TOP was defunded in the FY 2005 budget. Sascha Meinrath, who noted this defunding on his blog, said via email that an important ally of community wireless groups was lost with the end of this program. The program disbursed $250 million since 1994.
You see him here, you see him there, you see him Wi-Fi everywhere: Nigel Ballard is Portland's Wi-Fi king: Nigel doesn't offer service himself, but he's behind or beside--with many other volunteers--the most public and popular free Wi-Fi installations in Portland, Oregon. He's also a commercial Wi-Fi guy at his day job, in which capacity he installs service at resorts, golf courses, marinas, and, yes, hotels and motels. Captive venues, let's say.
Nigel wrote this linked analysis at the always interesting MuniWireless.com. He notes that truly captive hotspots can charge $10 a day (or more, I've found). They have their audience where they want them, and unless they're non-exclusive--like having a choice of similarly priced hotels--they don't have competition for those spaces. (Well, not yet anyway: as 3G spreads, people may opt to spend $20 to $80 per month for unlimited 100 to 400 Kbps service instead of $10 per day for captive higher-speed access. Depends on the applications.)
For more competitive locations, like coffeeshops, Nigel argues it's a race to the bottom to charge nothing for service. And he has numbers and years of service to prove his contention. Personal Telco Project (PTP) gets a venue set up for about $60: they buy a router and use a donated PC that Personal Telco has rehabilitated. The folks at PTP recommend business DSL, about $45 a month, which gets around the shared-access issue.
Nigel also makes a good case for why T-Mobile's plan of unwiring Starbucks makes sense: ubiquity and consistency. This is a message I've put out for years, too. If you need high-speed--there's a T-1 in each location--and you want an almost certain chance the service will work and be available in many places wherever you travel, T-Mobile is the service for you.
But Nigel addresses the PTP "customer": ...you just want a good cup of joe, a comfy chair and free access to a Wi-Fi node, then the community model may well serve you and your wallet better.
From the financial angle of the venue, Nigel points to World Cup Coffee's experience in trying to charge for Wi-Fi with Toshiba SurfHere's service. They lost customers to PTP nodes nearby. Turning off SurfHere had results he doesn't report--they're obviously happier--but they only had a handful of customers pay for service off months of having it on a fee basis.
Wading into more challenging waters, Nigel points to SBC's $1.99 per month unlimited FreedomLink Wi-Fi hotspot service that SBC introduced recently to SBC DSL customers. I believe he's incorrect about the billing and customer service costs: this is incremental to the existing overhead of servicing DSL customers, which already pays the associated expense for those factors. $1.99 is just more revenue.
Nigel wonders where this rate leaves Wayport, T-Mobile, and Boingo. In pretty good condition, I'd wager. Wayport has already turned their ship, moving from per-session fees to per-month-per-location fees. If an aggregator wants to roam onto McDonald's or, soon, Hertz locations operated by Wayport, that aggregators pays a fixed amount per month for access to those locations. They don't have to recover a per-usage fee, which means that they're free to offer it, as SBC is, to millions of customers without increasing their cost basis.
Wayport will eventually try to shift all of its contracted hotspots to the Wi-Fi World model, they said months ago, and they'll have the SBC usage data to show hotels and airports why it makes sense. Airports might be hard to convince because they're truly captive outside of cellular. (And they make money by charging cell operators to have good reception in the airport terminals.)
T-Mobile still has the factor of ubiquity now coupled with a greater international presence. Wi-Fi is not racing to free outside the U.S., and that's part of T-Mobile's worldwide advantage. Because T-Mobile just added 802.1X authentication, that gives them an additional leg up: T-1, business-grade security, high reliability, and per-user unique encryption over the local link. It's an IT person's dream, and at $20 per month unlimited use or per-session resale through iPass, it's not a sufficiently high bar to cross for road warriors and their companies' expense-approving managers.
Boingo Wireless is only nominally a customer-facing service, and until free is really ubiquitous, Boingo's client software is the best thing out there. At some point, SBC's portfolio of hotspots might be better than Boingo's, but Boingo might also be reselling SBC and Wayport Wi-Fi World locations when that point would be reached. At $22 per month for unlimited usage worldwide (for virtually all locations), Boingo is probably the best bet for an international traveler visiting countries that they serve. Boingo feeds out Wayport, which gives you inclusive access at many hotels and airports, too. Again, not a hard decision. In any case, Boingo has a large business building private-label and back-end software, so it's unclear if long-term their strategy is offer a retail price and retail brand.
I agree entirely with Nigel that free is making huge inroads, and most spectacularly in the hotel industry which I would find surprising if, outside of certain independent or boutique properties, breaks Internet service out as a separate fee within a year or two. Some hotels will definitely continue to try to charge, but as the majority of them offer free Internet and free Wi-Fi, the number crunchers will find that unless they try to compete solely on room rate, they can't make it work.
Wayport's Dan Lowden told me Monday that certain Wayport hotel properties now see 25 percent of guests using Internet service in a given night. Wayport operates both fee and free properties, including the Wyndham chain which offers free Internet to members of their no-cost guest club. Even if the highest usage is at Wyndham, it still shows that the Internet is a service that's being used at hotels, and thus the race to free is certainly going to continue there.
Some New York City residents are taking advantage of a project offered by NYCWireless that teaches them how to set up a hotspot to share with their neighbors: Neighbornode is a project offered by NYCWireless that includes a package of open-source software that helps interested folks get started. NYCWireless is also offering workshops. People who set up the hotspots are encouraged to set up electronic bulletin boards that allow users to communicate with each other. So far, some of the hotspots are getting people together, for social reasons as well as neighborhood watch concerns.
No mention here of whether NYCWireless offers advice on avoiding the wrath of broadband providers like Time Warner, which sent out letters to people in New York City a while back warning customers against sharing their Internet connections.
Time article on Vivato, wireless clouds, and community/metro wireless: This article is so uninformed to my eye that I refuse to comment on it--I mean, look at how he describes Austin's network--instead allowing the community of readers to this site their opportunity to post remarks. Click Comments below to contribute.
Michael Oh attended Making the Connection: The 2004 National Summit for Community Wireless Networks in August: Michael Oh is the fellow behind NewburyOpen.net, and the owner of tech superpowers, inc. He sent in this report:
First off, I think the biggest thing was simply that everyone got together. The CWN (Community Wireless Network) world is something that we're all very involved in, but we're all very locally focused. It's almost by definition that we're mainly interested in our own hometowns, but sometimes that keeps us from seeing that CWNs all end up having a lot of the same challenges.
So, I have to give props to Sascha Meinrath and the crew at U of I that put this together - it was something that we knew there was a need for, but no one had done it - and they created it out of thin air.
I'm also surprised that so many people showed up for the event. By my count, there were somewhere around 150 people, from people like Rob Flickenger (of Metrix and O'Reilly's Wireless Hacking book) to Michael Calabrese of the New America Foundation. More surprisingly, there were people there because their grants had funded them to come and do research on how wireless can help community development. That means that CWNs are getting a lot of exposure - and not just to the people that make them.
The turnout solidifies one thing - that CWNs are here to stay, and they are about a lot more than "competing" against for-pay wireless.
CMNs are about community, not about wireless. It's simple to say, but very complex to understand, since community is such a broad-based word. Still, this movement is one of the quickest forming community movements that I've ever seen or been a part of - and we all share the same idea.
That idea is that we all believe that community wireless will make a better world. We don't agree how, and we certainly don't agree what technology it will use, but we're pretty certain it will change how the world interacts. And this change is very different from how the corporations playing with WiFi in public spaces imagine.
The meetings were organized in three tracks - Organizational Models, Technology, and Policy. Going in, I thought that 2 out of 3 were of interest, but the Policy track was really kind of strange to me. I didn't understand fully how spectrum policy really would help the future of our cause.
This is one of the big benefits of the Summit - spending time with the policy "wonks" (which regardless of name, are very interesting people to hang around), at the very least because you're happy that SOMEONE is interested enough in spectrum policy to fight in the halls of Washington for us. Howard Feld and Michael Calabrese were people that I met, talked to, and understood after the summit. You may even see me in Washington sometime if they have their way... :)
Organizationally, the other benefit was meeting all of the other major CWN players in the country. There were people from NYC Wireless, Austin City Wireless Project, PersonalTelco, and others... We got to hang out, share stories of crazy wireless projects gone horribly wrong, and drink beer. I'm hoping that one of the results from the Summit is simply communication between all of the different cities, so we can all work together to further the cause.
I would love to coordinate wireless festivals in 4 major cities on the same day, share technology, or at the very least just have WiFi webcams that allow people in 4 coffee shops all over the country to say hi to each other. We would love to see projects like our Boston Music Project not only exist in 4 cities, but also be connected between the different places, so that anyone in a coffee shop in Boston could experience the newest local bands in Austin. (FYI - this idea was sparked by Rich from Austin's work)
It became obvious that CWNs were already diverging into ideas about culture, arts, media, and community content. While incredibly powerful in concept, we also risk losing the focus of people outside of the WiFi community with this divergence. If we focus on such broad-based ideas, we look a lot more like community non-profits and less like WiFi organizations. Unfortunately, "WiFi as commerce" is the baby of the business section, not "WiFi as community development." In my opinion, that's why all of the CWNs have "fallen off the map" compared to the T-mobiles and Tropos Networks of the world - they're just no longer interesting to the business community anymore.
But I predict that the next wave will be when CWNs begin to provide "national local" content providing interesting local content for their respective cities, but as part of a national movement. That's the vision that I have for CWNs going forward - it's just a matter of getting others to sign on.
Fact is, we're all volunteers, trying to feed ourselves with other activities, so the chances for success depend on how dedicated we can be to another cause when we're already stretched for time. Luckily, there are already great examples of how this happens.
One of the best connections that I made was Prometheus Radio, the guys that came from pirate radio and now specialize in 'powering up' Low Power FM (LPFM) stations. I participated in the Portsmouth, NH, LPFM barnraising last month, and they sure know how to motivate people. FM people have been fighting the fight for many decades - pushing back against the incumbents, correcting FCC shortsightedness, and organizing disparate groups all over the country. CWNs could learn a thing or two from them.
I think that if organized well, we can make the community wireless movement more than just an annoyance to the T-mobiles of the world - by providing content that can only be found locally. And the 1st Summit was a huge step towards that - even if it was just getting everyone in the same room.
Vancouver, Washington, gets its first free public Wi-Fi for a total of over $30,000 and after a year of planning: Vancouver is a popular town to live in for folks who like to shop just across the Columbia River in sales-tax-free Oregon and live in income-tax-free Washington. It's also a less-bustling town than its neighbor, Portland, edging right into undeveloped countyside, even as it's become a sprawl-y bedroom community.
While Personal Telco in Portland has what they report as over 100 active nodes in their free network, including prominent ones in Pioneer Courthouse Square in downtown, Vancouver's first free public space Wi-Fi appears to be a 3 Mbps Internet feed over two antennas relayed from a nearby building into Esther Short Park.
The project involved $30,000 from HP in goods and services as a grant, which seems improbable just to start with, since the coverage area and relaying involves very little engineering and could use off-the-shelf equipment. It would be hard to spend more than a few thousand dollars at commercial rates and using enterprise hardware to achieve the project's scope. But HP didn't install the equipment: that was done at a reduced rate by Vancouver Power Systems. Bandwidth and authentication services are being donated by local companies. The equipment is 802.11b, not 802.11g, even.
The nature of outdoor Wi-Fi seems to be profoundly misunderstood by either the reporter, the officials and others talking about the project, or both:
Patrick Gilbride, the city’s information technology manager and main liaison to the project...[said] “If this is successful, I think it will be attractive to look at providing a service that helps local businesses on a wider scale.” To do so would require a much larger investment in equipment – or a much stronger signal than Wi-Fi, whose range is generally limited to a few hundred feet. Gilbride pointed with enthusiasm to the coming next-generation of wireless, known as WiMax, which promises an exponentially larger coverage area than Wi-Fi. “WiMax could be huge, but we’ll have to wait a year or two before it’s really available,” Gilbride said.
In outdoor, public spaces, Wi-Fi can easily span several hundred feet unimpeded, and more with a little antenna design or a small cluster of dumb access points (as little as $50 each) with sectorized antennas. WiMax might help on the backhaul, but moving 3 Mbps in any direct line of site can be accomplished for as little as a few hundred dollars without any special involvement. The WiMax flavor that would offer the coverage in question is probably 2007 to 2009 timeframe; the near-term WiMax is only point-to-point service, and similar technology is already available today, just without the brand and standardization. Mobile WiMax is currently a pipe dream that cellular data and Wi-Fi might render entirely moot, but we'll see.
I asked NIgel Ballard, one of the movers at Personal Telco (and a former employee of ElevenWireless, which donated its services for the Esther Short project), about the cost and timeframe of the Vancouver project. He wrote back:
It was a year in planning and consists of two, count 'em, two access points in total! I told the journalist [who reported this story] that I had a coffee shop on 21st avenue that had three access points, respectively running A, B, G and it took three hours to unwire!
While it's unfair to compare coffeeshops and public parks, Personal Telco has unwired larger spaces, as have groups around the country. In Salem, Mass., Michael Oh worked with a community group to raise $10,000 to build out three full streets of downtown access with an aggregate bandwidth of 18 Mbps (three 6 Mbps DSL lines, one for each street). The $10,000 covers the cost of survey, equipment, installation, and most of the cost of the first year of unsubsidized bandwidth.
Sascha Meinrath of CUWiN offers his follow-up on previous posts about mesh networking's scalability and utility: Continuing a conversation that began back here, and continued here, open-source and world-wide community mesh networking developer Sascha Meinrath replies and elaborates on those posts.
Chari is right on the mark with his clarifications on network performance degradation rates. The case I had made purposefully oversimplified the throughput degeneration rate. However, in real-world deployments, the actual throughput of a network probably degrades at somewhere between 1/n and (1/2)^n -- where n is the number of hops. Think of these two equations as two limits of the probable degradation rate; as anyone graphing these functions can see, they map an increasingly wide area of probable degradation rates as the number of hops increases -- representing an increasingly large "unknown". The point is that exact throughput degradation rates are fairly impossible to pin down because the variables that need to be taken into account differ by locale. As anyone who has done numerous real-world implementations will attest, bizarre confluences of factors can sometimes cause unanticipated outcomes and disruptions.
One of the major problems facing wireless deployers is that almost all research has been conducted either via computer simulations or in "in-vivo" deployments that are highly contrived (often within science buildings or even within single laboratories). This research provides extremely useful guidelines for anticipating problems; but often fails to capture the complexity of deployments in the community. A closer-to-life example of "real-world" usage is MIT's roofnet project, whose deployment is being used to help proof the ETX route prioritization metric that is being integrated into CUWiN's software. However, this network is utilized mainly by computer science students, who are not exactly representative of the population at-large.
Nitin Vaidya's work has made tremendous strides in our understanding of ad-hoc and multi-hop networks (which Chari does well to point out); but what is really needed is a truly community-based network (with all the attendant messiness) that can be utilized to explore the real-world limits of wireless networks. It is with this goal in mind that Nitin, David Young (CUWiN's technical lead), and I co-wrote an NSF grant proposal entitled, "Engineering Community Wireless Networks" earlier this year. For companies and entrepreneurs working on wireless networking solutions, the possibility of gaining real-world data is extremely valuable. Likewise, for those of us working on Community Wireless Networking solutions, these data will provide an opportunity for better understanding the constraints for deploying robust networks and create more precise parameters for degradation rates.
Chari was also right on the mark in pointing out that scalability and performance are not necessarily directly linked:
"Scalability speaks to how large of a network you can build. Scale is unrelated to the number of hops. A mesh network with a small number of nodes but few wired backhaul points and/or an in-line topographic layout may have a large number of hops. Conversely, a mesh network with a large number of nodes but many wired backhaul points and/or a lattice-style topographic layout may have a small number of hops throughout."
However, I would argue that they are still significantly correlated. In either case, an ideal networking system would be able to handle any of the topographies Chari alludes to as well as allow for multihoming (the use of bandwidth from multiple internet connection points for a single download or upload). Multihoming, however, brings us back to the same problem of scalability and hops being highly correlated. Most importantly, as Chari states, "The real limit to scalability in most mesh networks is routing overhead." Both TBRPF and OLSR protocols share the problem of not scaling to extremely large networks. They're built with a cell-phone, tower-based mesh topography in mind (one need look no further than the cute OLSR MPR flooding demo to see this). A-HSLS will scale to thousands of nodes arranged in a truly non-hierarchical fashion -- it's the difference between two protocols that are most useful by major telecoms, and one that is useful for community wireless networks. A-HSLS is more useful for Community Wireless Networking purposes, while TBRPF and OLSR are more useful for major telecoms.
As regards the 500mW limit I propose as a proactive solution in my manuscript, it is important to remember that one can today legally transmit at up to 1W; and with a exemption (which is pretty much a rubber-stamp process) and an amplifier, one can go up to 10W. The problem is not what power level one can transmit at with today's wireless card technology; but what will be rolled out in the future. I'm especially concerned about the WiMax technologies being proposed that allow for higher transmit powers within the same frequencies as today's Wi-Fi systems -- see this link.
The take-home message is simply that there are multiple uncertainties within wireless technologies -- and it is not that these unknowns should be viewed as a barrier -- but that there are very few universally correct answers. Whether one is looking at throughput degradation or "the best" routing protocol, wireless is a nascent technology with an incredibly diverse set of possible implementations. It is impossible for those of us working with wireless technologies to fully understand all of the different available options, much less have answers to all the questions that are asked of us. But in debating the pros and cons of different solutions, I am hopeful that we'll increase our collective understanding of wireless technologies and creat additional opportunities for making the right choice for particular implementations.
The head of a mesh company argues mesh networks don't scale -- and one of the folks behind an open-source mesh software project examines the argument: MeshDynamics sells a multiple-radio solution for mesh networking, and the head of the firm wrote a brief article explaining why single-radio mesh networks can't work beyond a very small deployment. I asked Sascha Meinrath of the CUWiN project for his feedback on Francis daCosta's comments.
(After this item was posted, Jim Thompson contributed his own extensive thoughts about both daCosta and Sascha's statements, hence the "third" view of the revised title.)
While I do think that Francis daCosta brings up some potential pitfalls to wireless mesh networks, the doomsday picture he presents is based on a flawed understanding of how mesh networking topographies work. I'll explain below:
1- Radio is a shared medium and forces everyone to stay silent while one person holds the stage. Wired networks, on the other hand, can and do hold multiple simultaneous conversations.
2- In a single radio ad hoc mesh network, the best you can do is (1/2)^^n at each hop. So in a multi hop mesh network, the Max available bandwidth available to you degrades at the rate of 1/2, 1/4, 1/8. By the time you are 4 hops away the max you can get is 1/16 of the total available bandwidth.
This problem exists only when all tranceivers within a mesh topography "see" each other. And herein is the flaw in the argument. Within a mesh network Request To Sends (RTSs) do silence nodes within range; however this degradation moves in waves--so if part of a mesh consisted of 7 nodes (of which G is connected to the Internet):
A-----------------> ------B-----------------> ------------C-----------------> <-----------------D-----------------> <-----------------E-----------------> <-----------------F-----------------> <-----------------G-----------------> | Internet Connection
Here's what would happen. A would pass a packet to B; when B passed a packet to C, A couldn't talk--thus the 1/2 reduction in throughput; when C passed it to D, the same problem would occur for both A & B (thus a 1/4 throughput); likewise for D to E (because D would silence A, B, & C), thus a 1/8th throughput. However, when E passes a packet to F, A is unaffected, when F passes a packet to G, both A & B are unaffected. Thus, in this solution, throughput would theoretically max out at 1/8th (which is probably still much more throughput than the average Internet connection--where the usual bottleneck resides).
What this really points to is the need for power control in radios (which is something that CUWiN wants to work on), smart antennas, and other innovations that help to create wireless topographies where as few radios as possible "overlap." I've written about some of these solutions in a paper that I'm adapting for a book chapter -- you can download this.
3- That does not sound too bad when you are putting together a wireless sensor network with limited bandwidth and latency considerations. It is DISASTROUS if you wish to provide the level of latency/throughput people are accustomed to with their wired networks. Consider the case of just 10 client stations at each node of a 4 hop mesh network. The clients at the last rung will receive -at best- 1/(16,0000) of the total bandwidth at the root.
This simply points out the need to separate inter- and intra-nodal communications architectures--a problem that CUWiN has both already identified and implemented.
4- Why has this not been noticed as yet? Because first there are not a lot of mesh networks around and second, they have not been tested under high usage situations. Browsing and email don't count. Try video -- where both latency and bandwidth matter -- or VOIP where the bandwidth is a measly 64Kbps but where latency matters. Even in a simple 4 hop ad hoc mesh network with 10 clients, VOIP phones wont work well beyond the first or second hop -- the latency and jitter caused by CSMA/CA contention windows (how wireless systems avoid collisions) will be unbearable.
I do agree that QoS problems continue to plague most mesh wireless networks. It's a problem that needs to be solved and that most deployments and commercial (and open source) solutions sidestep. I think Francis is wise to blow the whistle on this deployment problem; I think that many commercial mesh systems have been way oversold--which will only make the problem worse.
I am constantly amazed at how little most wireless companies know about the physics, software, and hardware of the networks they deploy. Most don't even realize that if they're using routing protocols that use Standard Link State they're going to crash and burn when they scale up. For a quick graphic of the problem, just check out page 29 (labeled page 26) of this link.
This is why CUWiN is creating an A-HSLS (Adaptive Hazy Sighted Link State) protocol (as far as we know, the only open source A-HSLS protocol). We believe that routing overhead will kill networks well before throughput does.
I am optimistic that solutions will be forthcoming. What we really need today are "altruistic venture capitalists"--folks who are interested in investing in the public good -- people who will sopport the development of CUWiN (or other open-source projects that are working on these solutions) so that we can build mesh wireless systems that not only work and scale, but exceed our current expectations of what we, today, believe is possible.