I had the chance to work for a couple of months with my friend Mike Dawson last year in Afghanistan. He is the core person behind the OLPC project in Afghanistan and pushes for Free and Open Technologies in Afghanistan. Solutions like LXDE with it focus on a lightweight and energy saving desktop or decentralised low powered networks like Freifunk offer new opportunities to give people access even in remote areas.
As part of the OLPC project Freifunk networks were already deployed in five Afghan cities including Jalalabad and Kandahar. Regularly updated local servers – easy to administer small netbooks – in the local networks give people access to copies of many local news resources, Wikipedia and thousands of educational books.
I regularly receive news from Mike and I would like to share them with you.
We’ve successfully tested here in Afghanistan using Freifunk to mesh routers between classrooms so that we can avoid the need for doing ethernet cabling in the school. Now with the 802.11n hardware out there that supports dual band MIMO 2.4GHZ and 5GHZ I’m hoping that we can achieve a wireless backbone performance equal or better to cabled.
Some Freifunkers out in Italy managed to get 80Mbps over a 4KM link even:
q=cache:Ul-NcC60_tYJ:global. freifunk.net/%3Fq%3Ditem/open_ source_802_11n_big_ breakthroughs_are_coming+ freifunk+802.11n&cd=1&hl=en& ct=clnk&gl=uk
As far as I can tell 100Mbps (hopefully we can achieve 200-300) should be a reasonable throughput for the backbone for running the local library service / jabber / journal backup with about 600 laptops in the school, any opinions on that?
I was thinking of making a small transparent plastic container for it so that each one would sit slightly above the roof of each classroom, then connect to a normal 802.11g router in the classroom.
As per other deployments we cut the signal strength inside the
classroom; the classrooms are running on two non overlapping channels and the mesh backbone on another. We should have results by the end of the month. Given the cost of sending technicians to do cabling, feeding them, etc. I’m hoping this works out as about the same cost but more scalable.