If you haven’t heard about it, Elevate is a successful and extremely well organized festival about contemporary music, arts and political discourse. Each year, the festival picks an overall theme that accurately captures the most pressing topics of our time. This time, that theme was “The Apocalypse”, and festival activities revolved around the question of whether we as a species will be able to cope with a growing number of seemingly insurmountable challenges that threaten us, such as economic crises, failing governments, or ecological destruction. In answering this question, one central consideration was whether our current global political, economic and social systems can even provide the tools to address such problems.

Without falling back into a cyber-utopian perspective of the early Internet days, it is still true that the way we communicate with each other may play an important role in our ability to address our planet’s challenges. Finding the right balances in the online world between economic value and civil society use, between verified identity and anonymity, between individualism and communities, etc. is still key to our quest of using the Internet and related technologies for positive change.

The FreedomBoxing workshop had about 20 international participants, some with rather extensive prior knowledge about the FreedomBox project, and some newcomers who were interested in discovering what it was all about. We had quite a bit of hardware to experiment with, including 2 Dreamplugs, 2 Raspberry Pis, several Guruplugs, and networking equipment. We successfully set up a “Freedombox” Wifi network via a Dreamplug and its freedombox-privoxy service, through which workshop participants could surf the Internet free of certain advertising and tracking technologies. We then went on to demonstrate how a FreedomBox could act as a “personal data store” and run Unhosted apps, which means that you use applications on the web the way you always do it, however with the difference that your data remains on your box. One such Unhosted app is Friends#Unhosted, a decentralized social networking service. Several participants who brought laptops or smartphones were able to try it, and people were able to establish “friend” links and share posts with each other, while all this data was stored on one of our Dreamplugs.

My impression was that everybody understood and identified with the general idea, i.e. that we are working on a vision in which our personal data is not stored in the cloud and our communication is not handled by (often unknown) intermediate parties, but is rather kept on a box in our home and therefore fundamentally under our control. There were many questions during the workshop, and we had great discussions, switching between German and English language as needed. Finally, ten minutes before the workshop ended, as a highlight and to everybody’s surprise, a member of the FreedomBox Foundation‘s advisory board walked into the room and was able to provide us with some first-hand information about the current state of the project and about the foundation’s thinking on some of its technical developments. In total, I felt content that the workshop was a success.

Unfortunately I was only able to partially attend Elevate, but I was still impressed by the atmosphere of this event. Considering the theme of the “Apocalypse”, at times there was a sense of pessimism and hopelessness. However, and this is also my main takeaway, the prevailing spirit was one of an optimistic view into the future. One panelist mentioned that we have to embrace “courageous resistance”, another that we have to move “beyond fear”. And instead of tackling problems with “incremental repairs”, we should rather develop “positive scenarios for the future” which we can work on with all our energy and determination. This will require much strength, but anyone who has been to Elevate has felt that we have this strength.

Looking forward to Elevate 2013.

See here for the workshop description:

http://datenspuren.de/2012/fahrplan/track/Workshop/5003.de.html

You can find my slides here:

http://www.slideshare.net/peacekeeper/freedombox

And a video recording:

Overall, there was a great atmosphere at the event, and I met many interesting people. I was excited to learn that in Dresden there is also a community experimenting with the FreedomBox.

See here for a summary of the talk:

https://www.unwatched.org/DNP12_FreedomBox_Bedeutung_fuer_Oesterreich_und_Europa

And here are audio and video recordings:

https://dnp12.unwatched.org/Die_FreedomBox_und_ihre_Bedeutung_fuer_Oesterreich_und_Europa

(both in German)

I also had a live FreedomBox 0.1 at the conference, broadcasting a Privoxy-filtered WiFi signal which participants could use to surf the web without ads and online tracking.

I participated with the intent to work more on the FreedomBox.

I had the following concrete goals:

Goal 1: Build FreedomBox Image

FreedomBox images are regularly being published on the FreedomBox mailing list. These images can be flashed onto a plug server such as the Dreamplug.

I also wanted to be able to build the FreedomBox image myself, using the freedom-maker script. Existing instructions on how to do this are actually very good, see e.g. this page on the FreedomBox website, or this README file in the freedom-maker repository.

Even though theoretically this is quite easy, I encountered the following challenges:

1. I tried it on a clean Debian Squeeze, and ran into all sorts of errors with freedom-maker that I couldn’t fix. So I then tried it on the most recent Debian Wheezy, and that worked better.
2. I had to manually copy the kernel/uInitrd to the image, freedom-maker didn’t do it.
3. In the bin/finalize script, I had to change “projects-chroot.sh” to “projects-chroot”. Update: This has been fixed now.
4. Contrary to the instructions, I had to make sure my partitions were NOT mounted when I ran make, because freedom-maker mounted them by itself.
5. At the end, one mount (build/dreamplug/var/cache/apt) wasn’t unmounted. I have to unmount it manually if I want to run freedom-maker again.

Despite these minor observations, I was able to build a FreedomBox image, yay!

The output of the freedom-maker script had 72639 lines of text, I’m attaching it here.

Goal 2: FreedomBox on Guruplugs

I wanted to make the FreedomBox image work on my 4 Guruplugs. The challenge here was that by now the image was about 700MB – too much for the Guruplug’s 512MB NAND Flash.

My initial plan was to simply try to work on the image in order to reduce its size and fit it on the Guruplug. However, I soon decided that this was probably wasted time, since I would have to re-do the process again every time I wanted to use a new image.

The alternative was to simply boot the Guruplugs not from NAND, but from a USB drive or via TFTP. This is relatively easy with the U-Boot bootloader and involved the following U-Boot commands:

usb start
fatload usb 0:1 0x800000 uImage
fatload usb 0:1 0x1100000 uInitrd
setenv bootargs root=/dev/sda2
bootm 0x800000 0x1100000

Result: Success!

## Booting kernel from Legacy Image at 00800000 ...
   Image Name:   Debian kernel 3.0.0-kirkwood
   Image Type:   ARM Linux Kernel Image (uncompressed)
   Data Size:    2733148 Bytes = 2.6 MiB
   Load Address: 00008000
   Entry Point:  00008000
   Verifying Checksum ... OK
## Loading init Ramdisk from Legacy Image at 01100000 ...
   Image Name:   Debian ramdisk 3.0.0-kirkwood
   Image Type:   ARM Linux RAMDisk Image (gzip compressed)
   Data Size:    2455604 Bytes = 2.3 MiB
   Load Address: 00000000
   Entry Point:  00000000
   Verifying Checksum ... OK
   Loading Kernel Image ... OK
OK

Starting kernel ...

Uncompressing Linux... done, booting the kernel.

Debian GNU/Linux wheezy/sid freedombox ttyS0

freedombox login:

Goal 3: VirtualBox

The third idea was to run the FreedomBox image with VirtualBox. This had been discussed on the FreedomBox mailing list for a while, the intention being that it should be super-easy for non-Linux users to try the FreedomBox software on their own computer.

The problem here is that the FreedomBox software is for ARM architectures, but VirtualBox is a “virtualizer for x86 hardware”, not a hardware emulator.

One theoretical option would be to virtualize a Linux system in VirtualBox, and then within that, use the QEMU emulator to run FreedomBox.

However, early on during the hackathon I was persuaded by a fellow participant that this would be slow and complicated, cause all sorts of problems, and be generally not worth doing.

In addition, whatever procedure would be required to achieve this would have to be repeated with every new FreedomBox image.

Goal 4: FreedomBox and OLSR

My most interesting (yet as it turned out easiest) goal was to install OLSRd on my 4 Guruplugs, in order to establish a mesh network between them.

The reason for exploring this goal is that it would support a vision in which Internet freedom and independence are achieved not only with a personal FreedomBox, but also with decentralized connectivity, i.e. with mesh networks such as the Austrian FunkFeuer, or the ones envisioned by the Free Network Foundation. At a time in which highly centralized Internet structures on all layers are being increasingly criticized, the number of efforts to move to decentralized alternatives is growing, as is public awareness of the merits of such alternatives. Thus, an exciting scenario would be to set up FreedomBox’es as nodes in a mesh network such as FunkFeuer.

Making OLSRd work on the FreedomBox is the first step. Since I had already set up wireless ad-hoc networking in an earlier experiment involving B.A.T.M.A.N. (see this wiki page), switching to the OLSR protocol was extremely easy, and basically only involved installing the respective Debian package and configuring the correct network interface.

I ended up distributing the 4 Guruplugs in the Metalab hackerspace in a pattern that made sure not every node was able to talk to every other node. The final result is described in more detail on this wiki page and looked like this:

Other Impressions

Finally, I also spent some time with fellow participant Florian aka “uniqx”, who had also been looking into the FreedomBox.

Together we had the following ideas:

• We will try to make the FreedomBox image work on a Raspberry Pi, a topic which has also been debated a few times on the FreedomBox mailing list.
• He and a few other people are launching a great new effort called freie.it, which intends to provide free support services for free/open software and hardware. We agreed it would be great to offer FreedomBox-related support on this platform.
• Perhaps we will collectively order a few Dreamplug devices for interested individuals in the local community.
• Some time in August/September we will run a FreedomBox workshop at Metalab.

They are now calling for proposals for the Access Innovation Prize where you can win $20k.

So Michiel of Unhosted, Bjarni of PageKite, and myself have decided to submit a proposal, which would include building a simple FreedomBox prototype that runs an Unhosted “remoteStorage” component and PageKite to make it accessible from the open web. Also, the idea is to try integrate FreedomBox with the local FunkFeuer community mesh network in Vienna. We haven’t submitted the proposal yet (deadline is August 15th), but here’s the current text we’re working on:
http://projectdanube.pbworks.com/w/page/54796496/Access%20Innovation%20Prize%202012

Basically, the idea is that you could use any Unhosted-enabled web application out there, and your data remains on your FreedomBox.

If we win the prize, then this could serve a few purposes..

1. The three of us would have some $$$ to actively work and contribute to the FreedomBox at least for a little while.
2. The stuff we would work on (putting Unhosted and PageKite on the box) seems to align well with the “DropBox Replacement” idea that has been floating around.
3. We would have an actual (limited functionality, but working) FreedomBox, and a minimal viable product that can be demo’d at conferences.
4. The prize would mean a PR boost for the involved projects.

Working Breakfast on Internet Freedom

I attended this event on June 21st in Brussels. It was hosted by Freedom House and MEP Marietje Schaake, who has been active in promoting human rights online (e.g. see her video message to the Rio de Janeiro Human Rights & Technology Conference). The event gave an overview of the EU’s No-Disconnect-Strategy, an effort to support activists and promote Internet Freedom. A number of EU officials as well as online activists from various countries were present. Although it was agreed that participant names would be kept confidential, I believe I can share the following notes:

The No-Disconnect-Strategy has 4 pillars:

1. Development and deployment of technologies to promote Internet Freedom
2. Training and awareness for activists
3. Guidance for businesses about (anti-)censorship technologies
4. Development of European capabilities for situational awareness (monitoring the state of online censorship)

One EU Commission representative stated that civil society too often focuses on criticizing, and that it has a bad track record of making useful proposals. The question on how to fund the development of anti-censorship technologies came up. While the Seventh Framework Programme (FP7) is the EU’s primary instrument for research, it is also known to be complex and not easily accessible. One more flexible alternative might be the European Instrument for Democracy and Human Rights. Another question was the relationship of EU Internet Freedom programs with similar efforts in the US, such as the FreedomBox, or the Open Internet Tools Project. While participants of the meeting said that such coordination is happening and that projects should be complementary, it was also mentioned that European values are very different from American values, and that the EU should be leading in the field of Internet Freedom.

Digital Agenda Assembly – Workshop

One June 21st, I also participated in the DAA workshop about Social Media. There was a strong focus on economic aspects of social media services, and on the question on how they can help to create jobs and growth in the EU. Besides creating “digital jobs”, social media can also be used as business tools and lead to new economic models. One participant stated that Facebook was a walled garden and asked whether it would not be better to have social media that is more decentralized and pluralistic. The highly disappointing answer from the panel was that Facebook was not a walled garden, because everybody was on it anyway, and that therefore there were no walls between its users.

Digital Agenda Assembly – Plenary

June 22nd was the day of the plenary session. The video recordings can be found here.

Juliana Rotich – Executive Director of Ushahidi - gave a keynote speech. Some of the points she made were:

• The Internet provides “lateral” opportunities to get views “from the ground”.
• Data scientists will be needed to make sense of “big data” and do good for society.
• In Africa, mobile phones and mobile money are transformative powers.
• Personal data is everywhere, and it can be used for mining and inference. It can create new economic and social opportunities if the data is allowed to flow rather than be locked in, and regulation must not lose sight of this potential, however, there are also privacy concerns.

Besides giving an introduction of Ushahidi and enumerating a number of impressive success stories, she also touched on a few other projects and technologies, e.g. Global Voices, M-Pesa, Mxit, mySociety, and Personal. Two random quotes from her speech:

• “We need trailblazers, not gatekeepers.”
• “Talent is universal, but opportunity is not.”

A panel titled “Everybody online and empowered” followed. Some takeaways:

• Simply being online is not enough. Information must be 1. available, 2. accessible, and 3. usable. More software for inclusion of e.g. handicapped people is needed.
• Will we get to “Every European Digital”? We should be very ambitious.
• The right to be forgotten, as well as the right to not be online at all must be respected. Many 50+ citizens are terrified of new technologies.
• Digital skills of parliamentarians are important (“Nerds in Parliament”). There is a problem with politicians’ knowledge about technology.
• Democracy must find legitimacy in these changing times. It will only be strong if it also listens to minority voices. It is important to find and consider all opinions (those on Twitter are not everybody).

In the closing speech of the day, Robert Madelin of the EU Commission’s Directorate-General for Information Society and Media (“DG Connect”) noted that high-speed legislation is needed for a high-speed Europe, that the goal should be to connect more people more evenly, that innovative funding mechanisms should be developed, and that the Digital Agenda’s work must be personal, easy, and urgent.

I talked about how it can instantly improve your privacy online, how it can remove ads and tracking software, and how it can enable encrypted communication with other FreedomBoxes.

FM4 Connected – VPN, FreedomBox

(in German language)

The objective was for different distributed social networking projects to come together, understand each other’s objectives, share experiences, and align the different projects into a more unified effort.

My impression was that all participants agreed how such projects are becoming increasingly important, and how awareness of issues around online identity, personal data and privacy is steadily increasing among the general public.

The projects that were represented included GNUnet, SecuShare, Lorea, The Global Square, Briar, Deepamehta, UnlikeUs, Bitcoin, Tribler (online). Myself, I also tried to contribute by talking a bit about Project Danube and the FreedomBox. I brought the 4 Guruplugs which I had used earlier for demonstrations at the Internet Identity Workshop and the European Identity Conference.

There were a number of great presentations, both about concrete projects, and about general topics such as analyzing threat models.

My main takeaways from the event were:

• There was an amazing set of existing technologies, some of which are already very advanced and ready for use.
• The necessary effort to package up these technologies and turn them into an actually usable product seems to be much smaller than I had thought before.
• Some of the projects would align well with the FreedomBox.

My current favorite scenario is installing GNUNet and Secushare on a FreedomBox, and adding a pretty web interface that allows distributed social networking. I volunteered to help with work on such a web interface when GNUNet and SecuShare become ready. Yay.

Some notes of the event are here.

• Personal Data Ecosystem: Within this vision, a FreedomBox could act as a “Personal Data Store” (or “Personal Cloud” or “Life Management Platform”) which would enable individuals to keep personal data on a device within their home, and to exert choice and control over what happens with this personal data.
• Vendor Relationship Management: A self-hosted device such as the FreedomBox could also help to manage one’s relationship with vendors, and to express intents to purchase certain products or services.

There seems to be an obvious synergy, since the FreedomBox project as well as the two visions above all share the goal of giving individuals more independence and a better ability to control their communications online.

Judi Clark also brought one device, therefore we had 5 in total to experiment with.

Here are some pictures by Tom Brown of the table where the demo was held:

One picture by Jim Fenton of the agenda, see the top right corner for the FreedomBox session:

One picture by Tracy Sheridan showing myself after having received half a bag of chocolate in recognition of the FreedomBox demo:

And one by Doc Searls:

The scenario of the demo was as follows:

1. The Guruplugs were handed out to volunteer participants and plugged into power outlets.
2. Upon being plugged in, these small personal servers booted their Debian operating system and custom Project Danube demo software.
3. The volunteer participants of the demo were able to control their box via a web interface.
4. The first step to perform was to connect one’s box to the other boxes (using a button on the web interface).
5. The second step was to sign in to the network with an identifier (using a Distributed Hash Table), in order for boxes to be able to find each other.
6. After being connected and identified, the demo allowed participants to do the following:
7. Enter personal data which is stored in an XDI-based Personal Data Store on the box (first name, last name, email, etc.)
8. Establish a relationship with other participants, which allowed access to the personal data on their boxes via XDI Messaging.
9. Sending text messages from one box to another.
10. Sending an “intent” to all boxes on the network (via multicasting), indicating what one would be willing to buy at a given price.
11. Viewing “intents” received from the network.

The devices did NOT run any of the actual FreedomBox code, but rather some custom code I had developed within Project Danube. The demo was repeated several times over the course of the conference. I mentioned consistently that I was not directly affiliated with the FreedomBox Foundation, but that I had been observing it for a while and that the purpose of the demo was to capture the imagination of participants and provoke thoughts on how it could be used.

The demo involved a wireless access point to which all boxes connected. A mesh network using something like B.A.T.M.A.N. would have worked as well, but I found the Guruplug’s wireless hardware too unstable to rely on it for the demo. For some more information about using B.A.T.M.A.N. on Guruplugs, see here.

No authentication, authorization or other security measures worth mentioning took place in the demo software.

Many good ideas came up during the demos, for example:

• Stina Ehrensvärd of Yubico gave me a free YubiKey and YubiKey Nano, which immediately gave rise to a number of ideas of how this technology could be used for authenticating users in a peer-to-peer network. Since the Guruplug and similar devices have USB, these ideas made a lot of sense.
• Sam Curren of Kynetx had the idea that if such a box supported NFC, one’s smartphone could be used to assemble an initial configuration of the box, which could be installed through NFC contact. Alternatively, Bluetooth could be used too.
• Kelly Mackin of the Personal Data Journal explained that a peer-to-peer network on FreedomBox’es could be ideal for publishing and disseminating news items. Freenet comes to mind.
• One technology that has been described many times during IIW as having great potential for sharing and linking personal data is XDI. XDI can be used both in client/server models and in peer-to-peer systems.
• One challenge in any peer-to-peer network is the handling of events such as new connections, incoming messages, etc. In the demo software, this was simply achieved with small JavaScript snippets, however this seems like a hack. Perhaps KRL would be a perfect fit.

Many questions were also raised, for example regarding the hardware capabilities of the box. The Guruplug’s capabilities are of course limited, but even though it is one of the cheapest plug computers available, it was perfectly capable of running the demo software. By the time the FreedomBox or similar solutions become ready for the market, suitable hardware will also have become cheaper and more powerful.

The actual demo code consisted of an XDI-based Personal Data Store (for storing and accessing data), the Jetty web server, the FreePastry Distributed Hash Table, and a few other components.

On the IIW wiki, there is also a summary of what happened during this session.

Here are some screenshot of the Project Danube demo software that was used:

[Show as slideshow]

Start Page

Setting up a Connection to other boxes

Identifying to the network, so other boxes can find you

Storing personal data on one's box

Establishing a relation to another box

Viewing personal data on the other box

Sending messages between boxes

Creating an intent and sending it into the network

Receiving an intent from the network

During the workshop, I mentioned that a growing number of grassroots projects are working on completely distributed protocols for social networking and the sharing of personal data. I talked about the FreedomBox project, and showed a Guruplug to a few people throughout the conference. I suggested that the FreedomBox could play the role of what during the conference was referred to as a Life Management Platform. Although such grassroots projects are typically not a major focus of conferences like EIC, it did capture people’s attention and imagination.

The conference is clearly of a very corporate nature, mostly trying to explore the impact of identity and cloud technologies on business. However, there was actually a surprising amount of talk about the importance of decentralization. During the opening, one keynote speaker went as far as stating that “client/server” was really just a nicer term for “slave/master”.

Some of the concepts that received significant attention during the conference were:

• Life Management Platforms: Such platforms are a superset of Personal Data Stores and will not just store personal data, but also provide individuals with useful services based on it, enabled through mechanisms such as “controlled push” and “informed pull”.
• API Economy: With the rise of cloud computing, the importance of API-based communication between service providers will also increase.
  Several workshops about Identity and Security in cloud computing were held. It was mentioned that in the future, identity in the cloud will likely be based either on a monolithic identity provider holding all personal data, or on the idea of an Identity-Management-As-A-Service (IdMAAS) system that can assemble claims from different sources.
• Privacy-By-Design and Minimal Disclosure Technologies were also discussed, for example their possible application on smartcards or by governmental ID programmes in Australia, New Zealand, or by the European STORK interoperability project.
• Workshops on Key Internet Identity Protocols gave overviews of the current status of OpenID Connect, OAuth 2.0 and the idea of an Account Chooser, i.e. a standardized UI for web browsers.
• A session about Doc Searls’ book The Intention Economy explored how this new paradigm will unfold and what underlying infrastructure and business models it will require. PDEC Startup Circle member Connect.me announced the launch of the Respect Network and its Founding Vendor’s Program, which is aimed at taking the first step toward realizing the vision of the Intention Economy.

Thanks to KuppingerCole for the invitation, and congratulations to OpenID Connect for the award.