Notes from Broadband Bridge Community Wireless Meeting, Sept. 25 2012

There were twelve people at a Broadband Bridge meeting in Mount Pleasant to discuss the importance of community-controlled infrastructure for internet access.

First we read this news piece, “More sad news on why Americans pay so much for crappy Internet and phone service.”

People gave their reactions:

  • We should nationalize our telecommunications infrastructure
  • Other countries have free WiFi, why not us?
  • America has fallen behind over the last decade
  • We need uniform access, not just “available access”
  • It’s not an issue of cost – companies make record profits

Taking the notion of uniform access and nationalization, we shaped this into the idea of the internet as a commons. Points from this conversation:

  • Community access: concern for security
  • If I share, what if someone else uses it in a way I don’t like?
  • Liability is a tool that the ISPs use to force people to buy their own connection
  • Fear of liability comes from a lack of knowledge and understanding
  • How is that different? People don’t know the difference
  • Within the industry, the term “cloud” came from “not my problem”
  • Few people know that the cloud is all made of specific and fallible hardware
  • We could use the metric of “the weight of the internet” in electrons to ask, “how much does my part weigh?”

What are the problems with the internet and access in DC?

  • HacDC can’t get sufficient broadband for its projects, nor can other nonprofits in the St. Stephens space – this holds back the potential of the nonprofits
  • There is a chilling effect from the lack of trust of operators
  • A lack of alternative options prevents meaningful participation
  • With community wireless, you will likely be in contact with people who share their broadband, which provides real accountability
  • ISPs spend money on lobbying, not customer or community engagement
  • With the derecho – lost power, no internet even after the power came back, but community network had broadband bc the line for its gateway is buried
  • People want to work from home, but whether or not they have access to the internet makes their work from home a variable quality compared to working in an office
  • Are there places that have done community wireless before? (yes!)
  • We heard some negatives about mesh wireless, from this article.
  • Suggestion for organizing other networks: visit the ANCs
  • We are lost when we don’t think big enough
  • What has to occur, we have to say, we want an alternative instruction
  • We assume that we will fail when we don’t take control of our problems
  • Start by thinking big
  • Make the government work for us
  • We settle for less, then we don’t even get that
  • Pepco & Verizon aren’t supposed to determine our destinies
  • Classic situation: think both small and big at the same time

Then we discussed visioning: What is the best we could want for internet infrastructure and access in DC?

  • We want municipal pipelines and WiFi
  • Tech is isolative. People should understand the technology. The need for a large amount of education and practice means we should create learning beds, so people can increase their understanding.
  • We need something tangible here. We should generate a larger and a smaller vision, like let’s have access in parks and homes.
  • Every child will have access and hardware
  • Cheap municipal option for access that’s great, and then community mesh networks. OCTO needs to allow the community to connect, then we can have fast local access and a sense of community.
  • Reasonable municipal control of fiber in the ground, and totally free of censorship, surveillance, and self-identity is a choice.
  • We have total access available, but we need to work on the government to maintain good service for all – freedom to share resources
  • All communications services would be a utility, owned by the people, and government would provide physical infrastructure, system would be totally neutral – content ownership by originators, and abolish copyright & no software or business patents
  • Establish right for freedom to communicate, and let people know “you own this.” people don’t yet feel ownership. Also, moving into new centers of commons: change post offices into community centers before private developers scoop them up
  • DC-NET / DC-CAN is useful and shared to all, shared through community-controlled wireless mesh networks. Try to bridge a gap – have a huge mesh and good fiber. We must build mesh networks that make the idiotic dichotomy of access and the digital divide obvious.

Derecho Community Wireless

I came home from the Allied Media Conference to a DC melting in 100+ degree heat and a home without Internet access, courtesy of the now-infamous east coast derecho. Verizon provides our home Internet service, but they did not restore it until this past Thursday – five days after the storm. Happily we still had access thanks to the community wireless network, since the gateway in our part of the network uses Comcast. Comcast could have easily come down during the storm as well, but because the lines were separate, and on opposite sides of the street, people had a recourse. I noticed that the network had much less bandwidth available during Verizon’s outage. We never lost power either, thanks to the power lines being buried under the street.

Despite 70 mph winds and downed trees and limbs, that community-controlled infrastructure never went down. If MtPCWN already had direct chat running on the network, people could have easily logged on and started checking if everyone was ok – even if Comcast and Verizon were both down. As long as folks have Wi-Fi devices with batteries and the routers have power, we can communicate easily during a disaster.

At HacDC the other day, someone suggested that we could fortify some of the routers with backup battery supplies, specifically for use in disaster scenarios. This would be a good idea for a retrofit for a few key MtPCWN routers. When neighborhoods plan out community infrastructure, they could designate a certain number of routers – maybe 40% – as “critical infrastructure,” with the implication that they should be able to run for up to x hours, and connect to other critical infrastructure, in case of a power outage.

Treat It Like Infrastructure!

I had the great pleasure to speak to Nick Feamster‘s NOISELab group about Commotion and the Mount Pleasant Community Wireless Network while I was at Georgia Tech for ICTD 2012 last week. One of the Commotion slides shows a graphic of MIT’s RoofNet mesh network. When Nick saw this, he laughed – “I was a RoofNet node!” he exclaimed. “But people would just shove them next to their windows, or even dangle them out of windows so the routers hung only by the Ethernet cable.” Sometimes the network would go down for a week or more until the admins discovered that someone had a party over the weekend and a reveler knocked out a router’s cord from the power outlet.

Compare this to MtPCWN. In MtPCWN, all of the routers are mounted externally on the roof as high as possible, and we ensure that the power / Ethernet cord enters the building as securely as can be done with our resources. We even make sure that the PoE injector plugs into the wall close to the occupant’s existing Internet router, so if they ever choose to become a gateway, it won’t be a headache to connect their bandwidth to the public network. The hardware is all Ubiquiti and high-quality, and the silicone-filled Ethernet cable is designed for outdoor deployment. Meanwhile, the Broadband Bridge’s Bloomingdale network comprises almost no rooftop nodes, uses cheaper and weaker OM1Ps, and lacks any gateways right now – out of 26 routers in the network.

Preston Preparing the Ethernet Cable

I discussed these thoughts with Brian, who responded thusly: “If you want community-controlled infrastructure, you have to treat it like infrastructure.” That’s right. Our infrastructure must be appropriately priced, but if we use “cheap,” we’ll get “cheap” in return. If we think of community wireless as a community “hobby” or equate it with other “volunteer” efforts like cleaning up a park, then our expectations are the same as they are for other hobbies or half-day outings with pizza at the end. There’s plenty right with pizza at the end, but we should look to the Verizons and the FM radio stations of the world to understand how they manage their physical infrastructure. Towers are designed to stand for decades. Access to antennae is heavily restricted. Providers take great pains to place repeaters and broadcast towers atop of the highest ridges and the tallest buildings. Compare this with one of the Broadband Bridge network’s erstwhile gateways, a cafe. Their gateway Bridge router is placed haphazardly on the server side of their bar, an area that is not only high traffic, but experiences high and random throughput of dishes, liquids, mugs and plates. If we could take one hour to put it somewhere more out of reach and use cable staples to keep the cable out of line, most of the problem is solved.

Much of this is about control on the telco’s part, the problem we are trying to solve with community wireless networks. But we can’t associate everything about their infrastructure with their control. I think that one of the unexpected successes of MtPCWN so far is that because we treat it like infrastructure, it’s not in the way of people’s daily lives at the host locations. When the router is securely mounted on the roof and the cable doesn’t intrude on a resident’s regular passage, they can ignore it 99% of the time. That’s a good thing, because humans are humans, and we make mistakes. Keeping the devices out of sight and out of mind in this case is part of what makes community infrastructure human-focused. It’s accessible when we need it, but respects our ability to make a mess of things.

The Role of Safety in Community Wireless Networks

I have chatted with folks in my neighborhood, Mount Pleasant, for half a year about supporting the community wireless network I’m organizing here. The most common question they ask first is about “security” in an open network. It usually sounds just like this: “But is it secure?” And, painfully, they frequently interchange “open” and “secure” as regular as “work” for “job,” referring to a wireless network without passwords on the access points.

Recently, I thought about the values inherent in Commotion, an open source community wireless firmware project OTI is working on. Commotion will have built-in anonymization and security features. For all the importance of being secure from the start – security having new implications in the digital, networked world – the real potential of the network, the value growth, has to come from an intentional sharing and cooperation. Technical thinking about privacy though has got us in a general mood of securing against others, rather than actively reaching out. I suspect that this is the intent of ISPs, scaring everyone into locking down their wireless APs so no one will share their connection and everyone will have to buy their own Internet access. Plus some leftover hyper-individualistic attitudes from the Cold War.

This brings a new face to a problem of balance. I think that one part of the balance we seek when communicating – and one of the values that Commotion should stand for – is “safety.” Safety is a human term that’s understandable in its way to everyone, and yet we are likely to define it in different ways. For the whole communications network – infrastructure, software, the people involved – I think that “safety” is at the nexus of privacy, security, and awareness.

“Privacy” is keeping what you want to yourself – “to reveal yourself selectively,” to crib from Wikipedia. You have the right to “not be surprised” as Ashkan puts it. Your sense of others’ knowledge about your assets, information or affairs matters to your decisions and your well-being. Someone could sneak up to your window, or read information you send over a network, without you ever knowing – and without affecting the integrity of the information or your behavior.

“Security” is preventing your information, assets or affairs from being compromised or damaged. Unauthorized viewing of your data on a network is one thing, and having it scrambled, deleted, or secretly replaced is something else. Operation AntiSec, fueled by LulzSec and Anonymous hacktivists, brought this into the forefront of modern technology news. You have the right to not have your property damaged, of course, but the by-product of these attacks – whether by governments, corporations, or hacktivists of any stripe – is to make the digitally less-literate fearful of connecting and sharing. This is one of the things Commotion is meant to solve, with strong and reasonably sure encryption and anonymization by default. Which leads us to…

“Awareness,” the state of knowing what is going on in your environment and being reasonably confident of the consequences of your actions. Most people walk down the street reasonably confident they won’t be robbed, molested or attacked. If the odds were high enough you would be hurt, far fewer people would leave their houses and our society would suffer from the lack of voluntary connections. Security and privacy are great, but if people don’t understand what they mean, what is and is not protected, and what the likely consequences of their actions are on a network, then they are not going to be gregarious online. They will be recluses – society suffers again.

Bring these together and you have “safety,” something which I think all people can understand as a human right. Maslow’s hierarchy has it, we understand it to be a key human need on many levels, from freedom from violence to food security to knowing where tomorrow’s paycheck will come from. Technology needs to be built upon human needs, and so resolving those tensions – privacy versus sharing, security versus ease of use, awareness versus the vastness and complexity of networked communications technology – is paramount to developing a just, effective and enjoyable communications infrastructure.