One thing out of Egypt is pretty clear: our communications technology and infrastructure (mobile and web) are still dangerously subject to single point of control. Almost three years ago I explored a possible alternative with Phil Lamb, in a totally different context.

We were thinking about the problem of reaching subsistence rural farmers and we got into the challenges of distributing mobile phones -- namely, the towers necessary to provide the service coverage. So we began to wonder if there was a way to build a mobile network without towers -- essentially through the phones. The model for this is something used in wifi / wireless internet - "mesh networks" - where each computer on the wifi signal acts as a repeater for the signal. The One Laptop Per Child folks have implemented mesh network technology - and found it works (more or less). But laptops aren't as compelling as mobile phones in many contexts -- laptops require a lot of power, and the communication needs to organize in situations like Egypt are arguably served pretty efficiently by text messaging.

Here's the example Phil and I came up with for rural connectivity: A farmer in a rural area, living more or less alone, has one of our specially designed mobile phones. He has no phone coverage or service. But he can use his phone to identify that he is interested in hearing about new agricultural ideas within a 50- mile radius of his village. Every time the farmer's phone is near another phone, it passes that phone the farmer's request for information. Soon every phone in the nearby village carries this request for information, and any time any single person leaves the village and encounters other phones, a data exchange occurs and anything matching that farmer's criteria is sent to the villager's phone. When the villager returns home and his phone synchs the new data out to all of the phones in the village, the farmer receives notification that there are new messages pertaining to agriculture, and finds out that someone in the region is using a new method of crop rotation. At this point, the farmer can decide to do either send a message to that person asking for more data, or asking to meet and discuss his idea, or he can send his own message, describing a crop rotation method that is better still.

We call this idea "mailbags" -- essentially that every phone can carry a "bag of mail" from all the other phones it meets along the way. Ideally, these mailbags would basically become like servers in a datacenter, with the connections to those datacenters being provided by old fashioned foot power.

Taking the technology of mesh networks from the laptop world and applying it to mobile phones has tremendous possibility. Service providers aren't going to build them -- their recurring revenue stream would start to evaporate -- so we should. I know the Open Mesh Project has been working on this on the fly in Egypt -- but it's really time for a serious, organized project to create (essentially) mesh network text messaging devices.