Juergen's Background

From Gosh! 2009: Open Source Hardware Summit

Juergen Neumann

My engagement with wireless network technology started back in 2002, when I moved into a housing coop in East Berlin. At that time, in about half of Berlin, it was nearly impossible to get broadband access to the internet. I shared a telephone line with 35 people.

As an IT expert I thought about building a wireless link to an ISP about 1 km away. This worked quite well with the cheap off-the-shelf wireless equipment that one could buy by that time. Having solved my own problems this way, I soon realized that most of my neighbourhood had the same problem. So we started to link more and more people over the air to our wireless uplink. But, of course, the problem was even wider,and this problem led us to gather more people together to build a larger wireless infrastructure, and also a web community which we finally called freifunk.net-- meaning "free radio" in German. At the time, more or less the same scenario was occurring in many cities all over the western world.

Today freifunk.net is working together with many of these wireless communites, activists and FOSS developers all around the globe. Founded in 2003 freifunk.net has provided and gathered together the key developers of the OpenWRT project, a FOSS based firmware for wireless access points, and was in the lead of developing an open sourced version OLSR in the very beginning and later developed the b.a.t.m.a.n. mesh routing protocoll.

I have been a guest, an initiator and a co-initiator of many events and activities in the fields of the global wireless community network movement. Soon after starting freifunk.net, we organized and held the first global summit on free networks, titled "fresh air - free networks" [1][2]. The summit, held in Denmark in 2004, had over 200 visitors from more than 35 countries and 5 continents. We held workshops in DIY antenna building and taught people how to set up their own wireless infrastructures. One outcome of the summit was that we then started the "World Summits of Free Information Infrastructures" - WSFII [3]. Today WSFII gathers together many wireless community projects and activists from around the globe. Through WSFII, people have learned to build DIY wireless infrastructures in all parts of the world, giving their local communities a chance to take part in the local and global digital exchange. All of this was based on the principles of FOSS.

Understanding that wireless devices with meshing abilities would be key to provide ICT infrastructures (especially to rural areas in and outside Europe), we started to develop our own FOSS based firmware. The freifunk.firmware is based on OpenWRT [4], OLSR [5] and later also b.a.t.m.a.n. [6] and provides meshing capabilities that even today you can not buy from the leading brands for Wi-Fi equipment. Due to the great demand for more information about these technologies, a group of international activists wrote a free technical training book called "Wireless Networking in the Developing World" [7] that has been downloaded over 2.000.000 times. There is also an official guide from Africa's Meraka Institute that explains how to flash access points with the freifunk.net firmware [8]. This guide came out of the belief that Africa sees meshing wireless devices as a key to their recent and future ICT strategy.

WSFII activities had been widespread, so much so that in 2006 on invitation of a local community project and by warm words of the Dalai Lama, we went to Dharamsala, India and joined the AirJaldi Wireless Summit [9]. While in India I was also invited by the Indian ICT ministry in Deli to give a talk and presentation in front of the council, where I presented the decentralized freifunk.net community model and it's meshing technical solutions. The ICT's interest was triggered by the fact that they wanted to build decentralized ICT infrastructures in more than 500'000 villages.

During my time in India I realized two important things. First, an access points there is twice as expensive as here -over 100$- due to post-colonialistic distribution channels. Second, the task of re-flashing a router (i.e. an access point) is far too difficult for many people in the world. I concluded that we really need a cheap pre-flashed, meshing access points and a more direct method of distribution from the router manufacturers, mostly in Asia, to Africa, India and elsewhere.

Another observation was that product lifecycles in ICT are very short. FOSS developers all over the world spend a lot of time and energy by re-engineering digital devices. As companies don't typically provide FOSS developers with documentation of their layouts and the chipsets (i.e. since the hardware is not "open") developers have to hack most of the devices to make them work. Due to the fluctuating availability and the prices of chips on the global market, the next model comes out, and all the work spent, for example in OpenWRT, can be rendered useless as there is a new wireless chipset or any other change in design which is not yet supported.

To address this problem, we started a new initiative called the "Open Hardware Initiative" (OHI) [10]. Vic Hayes, the former chair of IEEE 802.11 and also known as the "Father of Wi-Fi" became our chairman [11]. We had a lot of good talks within this group and decided that we would need to get the manufacturers to pre-flash their routers with mesh-enabled FOSS firmware, and the need to convince them that they should only use parts that already are well supported by OpenWRT. Among this group was Michael Burmeister-Brown. He is one of the founders of Meraki [12], but quit them as they "sold-out" the open source idea. He had good experience in this field and enough financial background to get it going.

Our clear mission was to find a company who would build a cheap wireless access point as open as possible and pre-flashed with FOSS firmware and meshing capabilities, matching the needs of our target group- the many people out there in the world who are victims of the so called "digital-divide". Initially, we did not manage to get one of the big brands go for it. Therefore we (OHI) went to Taiwan ourselves and talked to various original equipment manufacturers (OEMs). In the end, it was Accton, one of Taiwan's leading OEMs, who is now building the first Wi-Fi router with a purely FOSS community developed firmware: the open-mesh router [13]. This is based on the Atheros reference design, which is not really free, but already very open. In the meantime, you can even buy other brands like Ubiquity at open-mesh.com that will also come preflashed with robin [14], an OpenWRT based zero-configuration FOSS firmware, mainly developed by Antonio Anselmi. The first lot of open-mesh routers was sold quite quickly and brought in the money for an optimized and even more open second edition.

Seeing the lack of understanding in regards to OSH in Asia, OHI also organized the first Open Technology Summit in Taipei last year. This was an important get-together of key developers in the Linux embedded scene, FOSS and OSH activists and Taiwan's OEM industry. While in Taipei the head of the Industrial Technology and Research Institute ITRI [15], Taiwan's leading national research organization, invited me to give a talk about the community-driven development of FOSS based wireless hardware devices to board of directors at the ITRI. Taiwan has quite an interest in FOSS-driven open hardware development for many reasons, as they see the global market move to China or even further, leaving them behind with a huge lack of understanding what they have been building for western brands over the past 30 years. In a sense, their experience is in the production of closed-source digital items, without learning anything about their technical designs.

In the meantime, a very recent project provides a very important innovation: a combination of a meshing Wi-Fi router, based on OpenWRT and b.a.t.m.a.n. combined with a PBX board and asterisk [16] for use with analog telephones as a SIP Voice-over-IP telephone client. (In other words: a low-power telephone and computer infrastructure for almost any place in the world.) Called the "Mesh Potato" [17] it is being built with the financial support of the Shuttlewoth Foundation and others. Here again Corinna Aichele, aka Elektra, who has also initiated the development of b.a.t.m.a.n. and who wrote the book "Mesh" [18], is one of the two leading developers. Elektra was one of the co-founders of the Berlin wireless community and later of freifunk.net. The second key developer of the Mesh Potato is David Rowe, who had also designed the Free Telephony Project, a free (as in speech) hardware and software for embedded telephony.

Last, but not least, I have been personal engaged in the wider field of wireless ICT. Together with Malcolm Matson, a UK-based early pioneer of IP based infrastructures and the original founder of COLT (City Of London Telecom), I co-initiated the OPLAN foundation [19]. Also, as I see a great need for more unlicensed wireless spectrum, I just recently co-founded the European Open Spectrum Alliance (OSA) [20]. The Open Spectrum Alliance is a coalition of companies, organizations and individuals working to unlock the potential benefits of bandwidth for all. We would be glad to see major companies and other organizations join that mission, because we think that this is the only way to gain the momentum we will need to free more parts of the electro-magnetic spectrum-- before the EU sells it to the major telecom monopolies.

Finally I think we should be clear about the fact that the more open our ideas and products are, and the more we give to world, the more we may get in return. Open Source is not necessarily a contradictory idea to that of commercial success. You can build a business on it as have learned from many open source projects and companies like RedHat or Novell. On the other hand there are many OS projects which only work for idealistic goals with no or little budget or simply because people want to create something that does not yet exist. All of these different players can cope together well with a very restrictive license like the GPL that protects everyone's input and grants open and free access to others people's modifications of ones own work.

I hope we will find something similar to work with OSH.

Shifting the synthesis of FOSS to OSH still brings up some serious caveats about open source hardware that need to be solved: licensing and costs jump to mind immediately. Mass production will always require huge amounts of money and imply ethical issues like the possibility of sweat-shop labour and environmental damage. On the other hand, low production costs are key to low product costs and to making things available for a broader part of the people in the world. It would be great if we could become clearer of what exactly we want, and what not, and if we managed to adjust the system to what we really want it to be.

That's about where I am right now.

Juergen

PS: You can find an english intro on freifunk.net here [21]

---

Personal tools

Juergen's Background

From Gosh! 2009: Open Source Hardware Summit

Views

Gosh!

OSH Research

wiki

Search

Create a book