When the Arab Spring took hold, protestors in the region discovered their most powerful weapon was neither a gun nor a charismatic leader. Instead, the lightening-fast, lassoing force of the Internet brought together hundreds of thousands. In response, fearful regimes cut off Internet access.
Volunteers from around the world sprung to the rescue of information-stranded activists. Groups began to devise ways to get demonstrators connected to each other and the Internet. One such organization is the Open Technology Initiative, part of the New American Foundation, a nonpartisan research group based in Washington, D.C.
The U.S. State Department awarded the group a US$2 million grant to develop innovative ways to sustain an Internet infrastructure in virtually any type of environment. Dubbed "Internet-in-a-Suitcase," by newspapers, the project basically aims to pioneer ways to make the Web available anywhere with added security layers, even in the event of an official shutdown.
Josh King is the lead technologist for Open Technology Initiative, and speaks to Arabic Knowledge@Wharton about how Internet-in-a-Suitcase works, and how his group is working to protect Internet freedom for people in the Middle East and domestically.
An edited transcript of the conversation follows.
Arabic Knowledge@Wharton: Can you explain what is Internet-in-a-Suitcase?
Josh King: First of all I’ll explain the context of how that phrase came about. It’s a term that was invented during the formulation of a New York Times article. We call it "Commotion Wireless." It’s a software project, not necessarily a particular piece of hardware. The whole idea of this project is to create a set of software packages that could be used on a variety of devices, in the event of a crisis where the Internet shut down and national borders closed. It’s software that can build an infrastructure out of devices that are already accessible in the area.
Arabic Knowledge@Wharton: So your group works more on the software that goes into laptops, cell phones, thumb drives, etc., to build an Internet infrastructure. Is that the idea?
King: Well, yes. Let me explain a little more of the history of the project and what it’s intended to do. The particular program at New America Foundation that we’re a part of is the Open Technology Initiative. The purpose of this program is to utilize and advocate the use of open-source hardware and software in society because we believe that it’s something that has immense potential for empowering individuals.
We’re also involved in projects in Philadelphia and Detroit to build out public computing centers and broadband adoption programs. To facilitate that project, we started developing a particular customized set of software based on all the stuff that the open-source community has been working on for a long time.
And then Egypt happened. We saw all these activists on the ground desperately needing tools to communicate with each other and with people outside of their country when the Internet was shut down. We were trying to think of what we can do to help those people on the ground. And lots of people all over the world were focused on the same issues. What we realized is they needed local infrastructure to allow them to communicate when the Internet is shut off. We also wanted to make it more secure so that activists under threat can have a reasonable degree of safety when utilizing these networks.
Then the State Department came along with their Internet Freedom Grant and we ended up getting the US$2 million grant for software development over one year to pool all the material from all the open-source projects that already exist and make it easier to use. Also, [we want to] make a huge body of documentation around that and create a secure and anonymous infrastructure on top of that. Since a lot of these networks are built on an open-source, community model, they aren’t necessarily built with security in mind.
Arabic Knowledge@Wharton: The grant is recent?
King: As of October, we’ve started to get access to the funding and where the one-year clock starts for the development of the project. We’re building off a lot of open-source technology that we already have. We don’t want it to be that it has to be a network engineer with a Ph.D. to put it altogether and get running, kind of how it is now.
Arabic Knowledge@Wharton: So you’re working on bringing the programming down to the level where an activist in Egypt or a street vendor in Tunisia can get Internet up in their country?
King: That’s right.
Arabic Knowledge@Wharton: Did you had any involvement in the Arab Spring when it first started?
King: We talked to activists in the Middle East and will continue to do so. Obviously, we want to create a project that meets their needs. Since we’re creating a lot of documentation, educational materials, and interfaces, we need to create something that’s accessible to people in the Middle East and all over the world. It means language translation but it also means a lot more. There are social and cultural contexts that influence how we interpret technical documentation and ideas on how to build networks. That’s even true among different communities in the U.S. We want to explain what it can do for a community of activists or residential communities. And we’re also building tools into the software interface to make it easier for multi-language interfaces, which is standard practice in a lot of graphical software.
Arabic Knowledge@Wharton: You mentioned that you were talking to activists in the Middle East. Were you able to help activists in Egypt, Syria and Libya? The Internet in these countries wasn’t permanently shut down.
King: We actually determined that in Egypt there wasn’t a technology issue. They didn’t have an Internet kill switch. The Internet is essentially made up of these autonomous systems. The government basically called up 37 or so providers so the Internet basically stopped routing traffic to and from Egypt. It was basically a series of phone calls to network providers that happened over the course of an hour.
As far as talking to the activists on the ground when that happened, we discussed different strategies with them. We sent preliminary software images that some people in the underground in Egypt or adjoining regions tried out and provided feedback. A major focus we’re putting in the one-year development cycle is putting in the security and anonymity that’s so very critical in situations like that.
Right now, I wouldn’t say our software is exactly safe to use, if I was an activist in an Arab Spring nation. It’s very much like browsing the open Internet. I would want to add more security to be safe so I wouldn’t be arrested. There’s no such thing as perfect security. There will always be dangers and ramifications.
Arabic Knowledge@Wharton: Do you have a specific list of the types of hardware that your software applications will work for?
King: We’re aiming for a broad range of devices, such as cell phones, wireless routers, laptops. Basically, the core transport medium for this infrastructure is WiFi. We’re aiming to support as many different devices that support WiFi.
For our purposes, we’re building on top of WiFi chiefly because it’s cheap and ubiquitous. You could deploy a network without having to ship out specialized equipment. It’s not like we’re sending out a magical suitcase that has network technology that could cost millions of dollars. Instead, we just create software that’s available for free and the devices that are already there can utilize it.
The whole aim of it is to create what’s called a mesh network. It uses WiFi to create a "device-as-infrastructure" network. A traditional WiFi model that you might have in your office or a café is a hub-and-spoke model. You have a bunch of clients that connect to a centralized access point, and that has a connection out to the Internet.
In this sort of model, there’s more of an ad-hoc network layout. Every device connects to any other device within range. And all of those devices connect to a bunch of other devices in range of them, and those devices connect to a bunch of other devices in range of them, and so on. So it creates a network where, if you could see the connections, it would look like a large spider web.
For example if you had a smartphone, and you’re at Tahrir Square, you’ll be able to connect to other people who are around or outside the Square who have smartphones. Even if the Internet is shut off, you can have services to allow people to communicate with each other that are completely separate from the Internet. So if you have an Internet connection, then that connection is made available to anyone on your local network. Instead of a satellite uplink only made available to a couple of people who have the satellite uplink, they can connect it to a device running Commotion and make it available to everyone on your local network.
Arabic Knowledge@Wharton: So you’re not really taking a suitcase and smuggling it across all these borders in the Middle East?
King: No, fortunately not. [Laughs] That sounds kind of dangerous. But we would be up for deploying networks domestically and abroad. And there are always going to be devices that are going to be more ideal for this type of mesh networking, so in some cases you might want to bring equipment along.
Arabic Knowledge@Wharton: This isn’t reliant on cell phone towers or anything like that?
King: No. Actually, one of the projects we’re working to integrate with is a project that does open GSM (Global System of Mobile Communications) cell phone networks. This means that if a cell phone network goes down, you can put up your own cell phone network. A GSM network has complicated infrastructure behind it. This basically has a GSM radio but behind it is the mesh network instead of the GSM network.
Arabic Knowledge@Wharton: So this will help when cell phones get shut down?
King: Yes, exactly. It’ll also make it possible for people who don’t have laptops and Smartphones to make use of their network with regular 2G "dumb phones." This is especially important in lots of regions all over the world where [simpler phones] are really the predominant method of communication. If we can provide some method for those phones to communicate using our network, that’s huge.
Arabic Knowledge@Wharton: Can this separate infrastructure ever be permanent or will it always be mobile and temporary to dodge potential saboteurs?
King: This technology is really flexible so it can support a huge range of deployment models. It supports permanent infrastructure, like the routers we have on rooftops in Detroit, and ad hoc support devices for people who happen to be in the same proximity as someone else with supported devices.
Arabic Knowledge@Wharton: So if you have footage of a protester getting beaten, you can jump that image from phone to phone to phone to get it out into the media?
King: Yes, exactly. That picture of the network that the device has is ever changing. If a node drops off the network, the network just routes around it and then finds new path between all the different points in the network.
Arabic Knowledge@Wharton: Have you met opposition to build this separate infrastructure?
King: Not particularly. There is a press release from Iran’s intelligence service saying they’ve already figured out how to defeat Internet-in-a-Suitcase. There are some indications that some governments and people are concerned.
Arabic Knowledge@Wharton: So when Iran’s intelligence system says they’ve already figured out how to defeat it, is that possible with flexible nodes jumping and moving? How can you really defeat that?
King: Well, since we mostly haven’t developed it, they’re a bit premature. We’re attempting to deal with all sorts of threats in a reasonable way. Nothing will be 100% secure and there are other direct methods, like jamming radio signals that could be difficult to completely combat. You can jam a network like this, but this network does have some advantages. This network uses WiFi that uses bands and signals that are used by whole lot of different things. If you jam it, it means you’re jamming every other wireless device in the area so there’s a larger penalty if you’re jamming those bands.
The devices are mostly low power, which means it can be difficult to detect unless you’re close to it, unlike a transmitter tower. The lengths are short-range so if you jam at a short range, then the network will just route around that area.
Arabic Knowledge@Wharton: This topic goes back to the digital divide issue that is going on in the U.S.; that every person should have a right to the Internet and information on the Internet.
King: That’s definitely a core tenet of the Open Technology Initiative. We’re all free software advocates in favor of open technology. It’s being run as a traditional open-source project so the entire process and source code and everything that goes into the project is freely available online and will always be. Anyone can download the software, take a look at the source code and modify it for free forever.
Getting software into the country once the Internet has been shut off isn’t going to be the easiest thing. The ideal situation is to have the software already in use, connecting communities in areas where, if the Internet gets shut off, they can flick a switch and build out a network. We want to deploy it to communities all over the world even when there’s no crisis situation.