Cool idea: Communal broadband


It's communal broadband, man.

A Boston company wants to offer people free access to a communal broadband network if a user pays for about $300 in equipment, according to an interesting summary over at Stop the Cap. The site profiled NetBlazr, which hopes to deliver free broadband with speeds of up to 60 Mbps to companies that will share that access with others in range while turning over management of the gear to NetBlazr. From the post:

NetBlazr starts with gigabit fiber from Cogent Communications, and then delivers free or low-cost access to any customer willing to do two things:

Spend $299 for the basic installation kit, which includes a high-speed router, three antennas, and some cabling;
Use the included equipment to receive service from NetBlazr and agree to share it with anyone in range of the wireless antennas included in the kit.
Reception of the wireless broadband signal, comparable to Business Class DSL, comes with no ongoing fees. If you want dedicated, guaranteed speeds, NetBlazr will sell them to you at an added cost. The more customers exchanging signals, the more robust and faster the network becomes, says NetBlazr CEO Jim Hanley.

This is a pretty cool idea (seriously go to Stop the Cap and watch the video), although it doesn’t look like it would help with the rural broadband problem. What’s cooler is Hanley’s idea that services like this could commoditize broadband. Has he talked to Google (s goog) yet?



It will be fun, while it lasts – but mesh simply can’t deliver real bandwidth, and with an all-you-can-eat-for-free model, it will be quickly saturated.

The sponsor will have to turn to a paid model, limiting “free” access to 15-minute spurts; the pay subscriptions will not be competitive with DSL, so only the aging hippies and corporate-bashers will hang on. It will end in bankruptcy, and the usual suspect will claim a conspiracy-against-the-people, rather than doing the math, looking themselves in the mirror and taking the blame.


Brough Turner

Fred, it’s not mesh and it’s not a consumer hotspot service. It’s a network of short range point to point links providing fixed Internet access, primarily focused on businesses in downtown urban areas. See some of my responses above for more info.


Using a “hub-and-spoke” topology with a little bit of free (limited) Wi-Fi access as a marketing tool/freebie, I think this could be done. Just using 2.4 Ghz (802.11b/g/n) for client access, and any other frequency (5.8ghz, 802.22, Wi-Max) as the back-haul could provide cheap device access anywhere you set the “Super AP” up.

Fortunately, Cogent is pretty good about providing back-haul access for dirt cheap. They did it in East Palo Alto for the entire area (

As long as there is a pay tier (sounds like there will be / already something set up), and tightly manage the free users (can be done very easily) then I think the ‘last mile’ problem is solved.


It won’t last beyond the first couple of years, equipment will start to fail and need replacing and people will complain what do you mean I need to upgrade or replace my equipment every couple of years to stay connected and realize there is no significant cost savings in the long run.

Brough Turner

Hippo, there is no doubt that equipment renewal must be part of the netBlazr incentives program. We’re actually more concerned with equipment upgrades (ideally every 30-36 months) than with failures. Once installed and working, most equipment such as ours becomes functionally obsolete before it fails.

Don Platon

My understanding is the user providing unlocked wireless access to the Internet is legally responsible for any illegal downloads by the community. FAIL


Like all the Starbucks and other free hotspots, right ? They can pop you out a disclaimer saying that by connecting you agree to do nothing illegal, otherwise you will be responsible not the hotspot provider.
It could work, no ?

Brough Turner

Don, netBlazr is not running hotspots. netBlazr delivers free and super-low-cost premium services to fixed locations that purchase and install netBlazr nodes. The actual participant gets a public IP address. netBlazr is acting as an ISP, conforms to DMCA requirements and is protected under the DMCA safe harbor provisions.


I think with the disclaimer this is something I might be interested in. That was immediately my thought though about illegal downloads. All of us I’m sure spend more than 300 bux on Internet in about a year or so so this could be an easy way to build out their network.


It looks like San Francisco’s Monkeybrains + FON. Nice combo indeed.


I’ve always said this. Let the communities and cities build their own fiber network, then let companies bid on providing service. Yes the consumer will have to shell out a bit more for equipment but isn’t it worth it for not being AT&T or Verizons bitch? Broadband will be so much cheaper. Once the big telecoms own the pipes you are basically screwed.


This is how the cellphone industry got it’s start. Amatuer radio (ham) put up repeaters and members were able to place radio to pots(plain old telephone) for the cost of the radio and maintanance of the repeater. You had to be licensed by the fcc to use the radio.
This is a beautiful step forward in broadband.
As for rural use it would work anywhere WISP is deployed.
Whitespace broadband will give coverage from any small town up to a sixty mile radius. Cost is currently unknown but could be inline with this companies cost.
A hybrid system of engenius ap’s in the 1 watt range for a hundred dollars could be built and run on solar very affordably too. With a good grid antenna say 24 or 27db ranges from high point to low points or other high points could be 10’s of miles too.


This isn’t a solution to a problem; it’s a Ponzi scheme. Bandwidth is expensive, even if you have a good fiber connection. So, the network can’t sustain itself unless it keeps using the profits from sales of equipment to new users to pay for the bandwidth for existing ones. Eventually, as the service saturates the market, it gets congested (because it can’t afford to keep up with bandwidth demand) and then collapses.

Tim Panton

That isn’t my understanding of their business model. The article doesn’t detail it , but from it is clearly a freemium model – folks buy equipment to extend the network and in return they get some (limited) free bandwidth. Others (hopefully enough) will take the premium service which pays for the rest.
The business is based on the observation that the difference in price of bandwidth varies hugely across a few city blocks, they are levelling that difference and making some money on the margin.

Brough Turner

Regarding Shannon’s law, nobody can violate Shannons law but you have to understand what Shannon’s law is. Shannon’s law establishes the maximum amount of information that can be sent over a continuous-time analog communications channel subject to Gaussian noise as a function of bandwidth and S/N ratio. You can work within these limits in a number of ways: more bandwidth; more signal power; less noise (directional antennas); more channels (MIMO) or characterization of the “noise” to the extent it is not Gaussian.

In many cases, netBlazr is using 40 MHz channels (versus 10 MHz channels typical of 3G and many 4G networks). netBlazr is using directional antennas (6-15 degrees versus 120 degree sectors for 3G or 4G) and short distances (50-200 meters versus 200-5000 meters for 3G/4G. netBlazr is also leveraging MIMO which brings diversity gain, spatial multiplexing gain and beamforming gain.

There are certainly technology issues to overcome, but netBlazr is not violating any physical laws!

Brough Turner

Tim is correct. This isn’t a Ponzi scheme — it’s a profitable business. The free service is faster than Business DSL but it’s a shared access system. The premium services are committed data rate services at ~1/5th the market rate, but still very profitable. The deal is, for either free or super-low-cost premium services, you first have to buy a node (3 radios and a router @ $300 and install it at your premises and turn management over to netBlazr). Thus netBlazr participants are paying to build the network. netBlazr is buying bandwidth at backbone prices and selling it at a profit to businesses that otherwise would be paying 30x to 100x backbone prices for their bandwidth (from the duopolists). The opportunity is there because US businesses are really being squeezed by the monopolists — much worse than US consumers.


Mesh networks experience so much congestion and interference that they cannot be faster than DSL. What’s more, the bandwidth costs are far higher than the value that can be added by a mesh node. It’s easy to demonstrate this on the back of a napkin. The business model isn’t scalable, sustainable, or profitable.


Posting actually in response to “BG” comment about meshing schemes. Single-radio mesh is not scalable beyond perhaps 500ft, which is why enterprise-level mesh implementations have multiple radios per node, e.g. one channel/band for the AP and another channel/band for backhaul. This approach can be further abstracted, e.g. 3 radios per super-node, to permit greater scaling, albeit at the cost of more expensive nodes and competing for free space on unlicensed bands. Also, speeds exceed DSL when you limit single-radio nodes to max 1hops. I can see 5Mbit/s+ sustained 1 hop out on cheap 2.4Ghz single-radio nodes, with 802.11n MIMO channels presently providing 40Mbit/s observed airspeed (of 150Mbit/s advertised), even when in IBSS mode, for backhaul.


Amazing how ignorant folks are of Shannon’s law. High order modulations won’t work on “dirty” unlicensed spectrum in dense urban areas. Again, laws of physics. Any paying customers they get will not be able to obtain high quality service and will be dissatisfied, while the freeloaders will kill the Ponzi business model. Watch it happen.


Again, you apparently do not understand the laws of physics or information theory. Shannon’s Law dictates that the above won’t work. They won’t keep paid subscribers because the service won’t be worth paying for, and the freeloaders will sink the Ponzi business model. I’ve been in the wireless business for 20 years and have seen all manner of harebrained attempts to defy simple math — the biggest bust among these being Ricochet, with Clearwire soon to follow. Don’t take my word; watch ’em collapse once they burn through their capital.


This is my last response to “BG,” since you’re putting many extremes in your arguments, making further debate kinda pointless. Yes, many radios on one channel will gobble up all available BW. However, your invocation of Shannon / Nyquist in such a dry manner should also imply that cellular phone service shouldn’t be financially tenable, yet it is. Physical limitations of the spectrum and signaling protocol are known, and generally impossible to ignore, even if I wanted to. To your specific points: bandwidth hogs can be handled are handled by throttling each client at the nearest node, so that their usage doesn’t eat up all bandwidth on the next hop and/or backhaul. Nodes can also limit the number of clients they accept, to help further alleviate congestion. Neither of these mechanisms are 100% effective for every worst case imaginable; mainly because the spectrum is shared and I don’t own it. Still, the motivation is to devise inexpensive broadband on a “best-effort” basis, which is already comparable in quality to what I get from customer-grade wired service providers in my area.

Brough Turner

netBlazr is not using “mesh” networking per se. I agree that conventional mesh networks do not do well for Internet access as the available bandwidth drops off x2 per node for the first few hops away from the Internet gatewy, eventually reaching an asymptote ~1/5-1/10th the bandwidth of the technology. With 802.11g, that asymptote could as low as 100Kbps.

ntBlazr uses multiple short-range (< 200 meters) point-to-point links and radios with highly directional antennas. We use directionality not to gain extra distance but to reduce interference. We turn down transmit power as much as possible to maintain ~100 Mbps over each link. We have a router at each relay point so we have fairly flexible control of where bandwidth is delivered. Even with directional antennas, one of my receiving points can see 23 different SSIDs, however 22 of those are down near the noise level while one (the paired netBlazr node) has 32 dB of margin. (All this is in the 5 GHz range where there is 455 MHz of license exempt spectrum available, i.e. more spectrum that the sum of what is owned by Verizon, AT&T, Sprint, T-Mobile & Clearwire). Yes, anybody can use this spectrum, but the wavelengths are short enough that highly directional antennas are feasible in relatively compact designs.

Brian Burwell

Bandwidth is dirt cheap! As in FREE! It’s the infrastructure that’s expensive. Otherwise there would be no home network market. Think about it; what is the most expensive part of your home network? It’s laying the cable. With the wireless mesh network the company is rolling out, the most expensive component is the wireless adapters its selling to customers. And mesh networks are much more resilient to congestion than cable internet networks. The choking point in a mesh network is the access point to the internet backbone (not the neighborhood node), but that’s easy enough to solve by adding another connection to it when it is needed.


Dontcha just love all this wishful thinking about broadband? Ever actually tried to purchase backbone bandwidth or set up a wireless network? Truth is, bandwidth costs big, good wireless hardware costs plenty and cannot be effectively set up by amateurs, mesh networks are easily congested even at low bandwidths, and customers whose access is not throttled or metered have no incentive to conserve bandwidth and will hog resources. Watch for this Ponzi scheme to collapse under its own weight within 6 months.

Brough Turner

BG, I have purchased backbone bandwidth in quantity. The issue in the US is that the backbone is extremely competitive wile the access networks are almost-unregulated monopolies.

In multi-Gbps volumes, at a backbone data center, Internet transit (95th percentile of 5 min measurements) can cost as little as $1/Gbps/month. It’s more expensive as you move away from the backbone.

In the US, about 3% of commercial buildings have competitive fiber – there are multiple ISPs that have their own dark fibers into the building. In these buildings (just over a hundred in Boston), bandwidth costs little more than it does at a backbone data center. In any other building (having only Verizon or Verizon & Comcast circuits), even one just 100 feet away, bandwidth costs are 30x to 40 x more.

A 30x price difference gives netBlazr some headroom to experiment with business models, incentives and community issues. We’re still in development, but I invite BG (and everyone else) to check back in six months.

Comments are closed.