Paul Gardner-Stephen & the Serval Project: Communications For All
by Chris McGivern & SF Team, 27 November 2019
Paul Gardner-Stephen is a researcher and Senior Lecturer at Flinders University College of Science and Engineering. His fellowship ran from 2011-12 and was spent working on the open-source Serval Project, testing and deploying mesh telephony and software to empower individuals and communities to access communications where infrastructure is unavailable or unaffordable.
The development of the Serval software progressed well, but soon met an overwhelming technological barrier. In 2012, wifi capacity on mobile phones was minimal and in order to make use of Serval, users would have to root their phones, thereby risk the loss of their warranties. Despite Paul making good progress in improving the accessibility of the software to more mobile phones, success in the short-term lay in the long-term - and expensive - task of either influencing mobile phone regulators and manufacturers, or finding another way around this technological impasse.
But Paul’s fellowship acts as an interesting bridge between the work we have supported in telecommunications, starting with Steve Song and continuing in later years with Peter Bloom and Luka Mustafa. We spoke with Paul to find out how things are going with the Serval Project today, and ask for his reflections on his fellowship experience. Below is a transcript of the interview, edited for clarity.
SF: Hi, Paul. Thanks for taking the time to speak with us. First of all, what was the inspiration for your work?
PGS: I’ve had an interest in communications since I was very young, specifically in digital communication. I look at how we enable communications in tricky situations without unnecessary costs to the end user. When the Haiti earthquake happened in 2010, I remember understanding what was going to happen with the loss of communications for more than two or three days. You get a complete breakdown of law and order.
I had this frustration of realising that everyone there had a mobile phone, but when the towers go down, you can’t use them. For commercial reasons we’ve never bothered to develop phones to do direct communication between one another. So that was the problem I set about to solve. To try and save human suffering for other folks in the future, so we wouldn’t have militia roadblocks, gangs in Haiti and the breakdown of order. Very early in the process, we were asking what happens when these technologies don’t exist, and how can we make things available when people need them.
And that idea turned into the Serval Project?
Yes. And while trying to solve this problem we discovered Steve Song and the Village Telco project. They had an interesting idea to try and fix telephony in remote locations with a mesh network. We realised that the same concept could actually work for mobiles, but it required someone to actually do it. It was through Steve that we found out about the Shuttleworth Foundation, too, so that was our introduction and we applied.
How did your fellowship go?
We were supported for a year to make progress on that and it was interesting on a number of fronts. The main thing we achieved was to take a crude proof of concept and turn it into a usable prototype. We also addressed some of the key things we knew needed to be done, particularly around security and privacy. We had the developers working full time over several months to reengineer all the protocols because we’d picked up something that wasn’t designed for mobile use. As it turned out, you could change a whole pile of things to actually make it work reliably.
The experience was great because we were able to take something that really was a half-baked idea at the beginning and through the support of the Foundation were we able to turn that into something we could continue to mature and find applicability in terms of helping folks in the field.
It was key to us getting support from other places down the track, as well. Prior to the fellowship, we only had something in the order of 10k through the entirety of the project. Since then we’ve had probably another million dollars from different places That was only possible because of what we achieved within the Fellowship.
Looking back, is there anything you would have done differently?
The fellowship starts very quickly. In my case, we had limited funds, and all of a sudden we had some amazing support. There is a sense of rush that comes with that. There were a couple of things we did that because we had to do them in that twelve-month timeline, whereas I think a better use of resources might have been if we had the equivalent of twelve months support spread over a longer period of time.
We put in a lot of effort getting traction with the 802.11 wi-fi standardisation stuff and we were making some good inroads on it. But it was an expensive process. I think if we had spread that involvement evened out over a longer time frame at the beginning we could have handled it in a different way, or engaged in a different way. But I’m not sure there is an effective solution for this, and every approach to doing things brings its own distortions.
I still remember the exit discussion and how their support is meant to make the difference between a project succeeding or failing. I thought it was a really interesting approach - and very sensible when you have limited funds - and it’s a patient approach. The desire is always there to make something happen as quickly as you can. The Foundation’s approach is that “it will end up where it ends up in due course, and we’ll delight in following its journey, but we won’t be rushed.” I think that’s the kind of decision-making process they go through. I would love to have more funding - always - but I totally understand the rational and appropriate way of spending the limited finances the Foundation has.
Did you take anything personally from your experience? And how did you find working with the Shuttleworth community?
Early on in the project we had been engaging with some of the other fellows and the big crowd of faces around the fellowship. That was really helpful to engage with those folks, meet them face to face and hear stories from the war trenches sufficiently related to what we were doing. It was really helpful and supportive.
By and large the process went quite well. It never felt heavy-handed and the team and fellows are never afraid of giving good advice. It’s very much “we think x, y, z. But if you want to do a, b, c, then we will support you in that.” It was a really nice aspect . It wasn’t that there was no accountability, but there was a very high degree of autonomy.
In retrospect, there were times where we took the advice and it turned out we were able to make considerable progress by doing it. That community of peers - not only the fellows, but the staff - it’s a real strength of the program. There are some very good people in that process. The team are excellent at focussing on the one thing that’s important, or using the experiences of others in their process.
It’s also grown my ability to look at other projects and interventions. There’s one we pitched early this year to make a secure solar powered super resilient mobile comms device but I wouldn’t be in a position to try and pursue that without the fellowship year. It was a real turning point.
How have things been going with Serval over the last few years?
The Serval Project is still alive and well, although not moving forward as quickly as it would have had we still got support. Much of our effort in recent years has gone towards developing our Mesh Extenders as a way to boost network range and work round the regulatory/handset problems. But it’s opening up interesting new doors. Along the way, we had a group of human rights defenders in Nigeria using the Serval mesh to communicate with themselves and Amnesty International use it to protect themselves against the local government. It was really interesting to see.
And over the years, we would know in advance where things were going wrong somewhere in the world. We had bursts of installs. For example when ISIS emerged in Iraq we saw a massive spike. We managed to find the forums where people were promoting it.
We knew something was going on in Crimea, as well. It turned out the Russians were jamming the Ukranian army’s communications - they were equipped with old Russian kit. So they couldn’t use mobile phones. Someone in the reserve army found out about what we were doing and actually set about building their own Serval mesh extender variant. It was actually being used by field artillery as their only means of communications. There are slides from the Ukranian army showing tanks and Serval mesh and how they can use that to defend themselves from this very asymmetric situation with Russia. It was totally unexpected.
More recently we have had support from the Australian government to run a pilot in Vanuatu as a disaster preparedness and response project. So there’s a dozen mesh extenders, a number of phones, and a lot of work to do over there.
Fantastic! And what are your hopes for the future?
We’re now at the point where we have working hardware, software and a list of critical bugs we need to deal with. But we are working with the Red Cross and they are looking to deploy our hardware in the next twelve months. They want to use what we created in the fellowship to change the way that they do field damage assessments digitally, and link that back through a resilient communications network wherever they are operating in the world. Its to support people reconnecting with families. In that sense it’s a really exciting time.
We’re also looking at how can we make a disaster early warning solution that is cheap enough to put in every village around the Pacific. We aren’t trying to optimise the commercial outcomes. It’s about coming up with a design that is cheap and reproducible, and tackles some interesting challenges, like lowering the cost of using satellites.
For example, I run a ground station for a satellite here on campus. We have a 4.5 million dollar dish on top of a building that wasn’t being used. Now it is, precisely so that we can create this system and do it cheaper. By us hosting it on campus it’s cut 20% off the cost of providing the low-band beam we are providing out there. So it’s been a transformation I have been happy with and it’s been very exciting.
That doesn’t mean I wouldn’t like to be rich and able to pursue more of these crazy things. But money is merely a means to an end, and my end is to leave the world a better place than I found it.
Thanks for your time, Paul. Any final words?
The Foundation serves a really interesting role in drawing people together of like minds to make interventions. And there were a whole lot of insights and personal growth that happened for me in the process as well, in terms of making me a much more effective social entrepreneur. For example - lessons learned - I have much more of a focus on product and understanding what is going to succeed in the public sphere and marketplace. And also being reminded - endlessly - to remain on focus was really helpful.
Another aspect is that I was quite comfortable with open source before the fellowship, but it has really strengthened my resolve and engagement to do everything on an open source basis. And also to do social entrepreneurship in an open source manner. In the same way the Foundation realised they didn’t need to fund me to change whether I would succeed or fail after that first year, it’s given me a similar philosophical outlook to the work that I do now.
If someone beats me to a particular social innovation or positive innovation, that is actually a positive. I can go off and try and solve the next problem that might come up. It’s the opposite of jealousy and greed. If I’m doing something to make a positive change and someone else picks it up and runs with it, great! If someone cuts your lunch as a social entrepreneur they have done you a favour - now you can start cooking dinner.
