An Interview with Stephen Song

Yelena Gyulkhandanyan:  When and how did the Mesh Potato come about?

Stephen Song: It came about after I joined the Shuttleworth foundation in 2008. I was aware of the potential of low cost wireless mesh technologies to create affordable infrastructure, but there seemed to be a challenge in getting these technologies to scale, and we had done some interesting pilot work, but nothing had really taken off. And so I convened a workshop in the middle of 2008 with some of the smartest wireless networking people I knew and so began to explore what were the key barriers. 

There seemed to be at least a couple of key barriers – one was that setting up a wireless mesh network was a complex procedure that required expertise. And second was that in many areas where we were interested in providing services, people were as interested in voice services as they were in data. Simply delivering data to a particular community, at least to rural communities anyway, seemed to be only solving half of the problem. So the result of that workshop was that we came to the realization, the conclusion, that what we needed was a hybrid of technologies, something that didn’t exist yet, which was a combination of voice and data technologies together.

We were lucky enough to have a brilliant open hardware designer from Australia attending the workshop almost by coincidence, and he said, “Well, why don’t we build our own?” Up until that point I think our dominant way of looking at the world was by asking what sort of North American or European technologies could we take and repurpose in Sub-Saharan Africa to address this issue of access in a more affordable way. The notion of actually manufacturing our own technology wasn’t on the chart at all and it took a little while for the idea to sink in, because it just seemed infeasible at the time. But sink in it did, which led through my fellowship at the Shuttleworth foundation to the funding of a pilot project to see whether it was feasible to complete at least a prototype design. The created prototype design led to a partnership with the manufacturer in Shenzhen, China, and to a short run of production which led to a bigger run of production. And so one thing led to another and now we have our own device that we manufacture.

Yelena Gyulkhandanyan: And how would you describe this device to a regular consumer?

Stephen Song: Well, it is a wireless networking device that works with similar units of its kind to form an autonomous wireless network that delivers voice and data services. So you can open a box of Mesh Potatoes, plug them all in, and instantly have a voice and data network. It is a network for which you don’t require a special voice technology. All you need to do to be able to start making calls is to plug in an ordinary phone into the Mesh Potato. So it doesn’t require any sort of additional smart VOIP hand set technology or anything like that. We deliberately chose to do that because analog handsets are very cheap and lots of people have them already or they cost less than $10 to buy. So it seemed like a very affordable way of creating a voice network.

Yelena Gyulkhandanyan: And how much does a Mesh Potato cost?

Stephen Song: They are about a $100 each.

Yelena Gyulkhandanyan: And how much does it cost to set up a network and what is the largest distance that it can cover?

Stephen Song: The cost of the network is literally just the cost of the Mesh Potatoes and so once you have them and they are powered up, you have network infrastructure that is yours for as long as the technology lasts, which should be many years. So that’s really the core cost; it’s just the cost of the devices. Then if you connect your network to the Internet or to the public switched telephone network you might have to pay for the access to the Internet or for access to voice services.

Each Mesh Potato has a range of about three to four hundred meters but the way the Mesh Potatoes work is each device acts as a repeater for the next one. So as long as the next house that you can see is less than three to four hundred meters away, you can actually build quite a large network, because if you have two houses that are six or seven hundred meters away, as long as you have one house in the middle that’s got a Mesh Potato, then all three of them are connected. Mesh networking has been around for a while but just hasn’t become as mainstream as WiFi hotspots.

Yelena Gyulkhandanyan: And in what frequency range does this technology operate in?

Stephen Song: It works in the 2.4GHz range which is your standard WiFi technology, which means that for most countries you can use it without requiring a spectrum license.

Yelena Gyulkhandanyan: So in what countries, other than South Africa, has this technology been deployed in? 

Stephen Song: Our biggest network is in the capital of East Timor in Dili. There is an NGO there called FONGTIL that has set up a large Village Telco network and there are a number of other smaller networks – one in Brazil, some networks in Nigeria and Cameroon, and then multiple other smaller more informal networks as opposed to formal Village Telcos.

Yelena Gyulkhandanyan: Have there been barriers in terms of deploying this technology? 

Stephen Song: A barrier for us is bringing the cost of manufacture down. So one of the downsides of being a very small organization is that in terms of negotiating with manufacturers and arranging deals we have very little leverage. So we will want to bring the cost of the Mesh Potatoes down by another 50 percent, which is completely feasible, but it’s a challenge to actually build the relationships with the manufacturers to get things done quickly.

Yelena Gyulkhandanyan: So what company currently manufactures this technology?

Stephen Song: A company called Atcom. 

Yelena Gyulkhandanyan: Can you provide a successful case study of this technology being deployed where it has made a difference in the village or where it helped create other social endeavors because people had access to this technology?

Stephen Song: Yeah, I think Dili in East Timor is probably the most successful example, in that the NGO that is running the network, FONGTIL, is kind of an umbrella organization for other NGOs in the region that need to connect and talk to each other on a regular basis. However mobile communication is quite expensive in Dili. So the NGOs have really valued being able to communicate easily and cheaply with their partner organizations through the Mesh Potato network. 

Yelena Gyulkhandanyan: Sounds good. Thank you very much for your time.

Stephen Song: All right, bye for now.