I got involved with OLPC back in June of 2006. It's hard to believe that was almost two years ago! I got involved for a lot of reasons but I have to admit I was tremendously excited about the technology in the XO. Over the last two years this enthusiasm waxed and waned. I first got to use an XO when Walter Bender shipped me a B-2 back in early 2007. I was tremendously excited but also afraid to mess with the little thing. It was far to precious to seriously tinker withk. We used it more to promote OLPC than to actually use as a computing or learning device.
From June to December 2007 I was almost 100% consumed with the problem of getting funding for Nepal's OLPC pilots at Bishwamitra and Bashuki schools. It was only in January of this year that I really got to use the little device. At OLPC offices I disassembled and reassembled two XO's. What fun! Probably the most fun I had had taking something apart since I tore apart an IBM compatible PC when I was 10. Taking the XO apart is a marvelous experience. You really see how amazingly well the hardware was designed.
Over the last two weeks I have been testing 130 XO's and two school servers ahead of our upcoming pilots at Bashuki and Bishwamitra. At times it has been extremely frustrating but also extremely rewarding. The presence service is extremely cool is somewhat unstable. The jabber server is a super cool technology but a pain to install. More and more I appreciate the Sugar UI and its design features. I also appreciate the incredible amount of work that the Sugar team puts into making Sugar a very exciting user experience.
The social aspects of Sugar are what make the XO so fun and so compelling. You just don't experience it with a few XO's in one place. David Cavallo is right when he says that the key to OLPC is saturation and integrating it into the fabric of education. I experienced this second aspect very clearly when I was at the OLPC Learning Conference in January. There were 25 participants from a variety of backgrounds, primarily from American school districts. We all had brought our own laptops to the conference and continued to use them as our primary devices through out the conference. Using the XO just wasn't as exciting as I had hoped. I contrast this with my experience at our 4 day teacher training program where the teachers used XO's as their primary medium throughout the program. The excitement and enthusiasm were palpable. I abandoned my own Dell laptop halfway through the program to work full-time with an XO.
The work this week . . .
Today we got 40 XO's connected to a generic wifi router, the Lantech WL54G BR, which roughly costs about $20 USD. All 40 were pinging the network gateway and none were disconnected. Then we tried to really saturate the networ. On 20 XO's we started streaming a 78 MB .ogg video of Rob McQueen's talk about Telepathy at Linux.conf.au. We were jubilant and amazed. Then we pushed harder. We got 25 XO's streaming the video, then 30, and then 37. At 30 the video stopped streaming on a number of the XO's but we were quite impressed.
This result also made us extremely happy because we had tested the Prolink WGR1004 which wouldn't allow more than 15 clients to associate. It's a frustrating process to procure equipment in Nepal because the local selection is quite limited.
Tomorrow we are going to add a second lantech to the office and see how well 70-80 XO's fare in dense environment. We will also do our best to crash Jabber using sharing. It should be a lot of fun. It's also great that we have two volunteers in our office helping us with QA, Prasoon Karmacharya and Ram Poudel.
I hadn't expected to still be so deeply involved in stress testing the XO network so close to our pilot. I expected to be working on Moodle, the E-Library, an online English-Nepali dictionary, etc. Still, I will be working on this project for the next couple of years here in Nepal so there should be plenty of time for that ;).
My thoughts on OLPC in 2008
I really feel that the general momentum around OLPC has slowed down. To me, this is a good thing. Expectations regarding this project were so atmospheric that nothing short of the outbreak world peace could have met them. The core folks behind OLPC -- both working for OLPC and in the community -- are going to create some great stuff this year that will wow everyone. 2008 will be a quiet year in terms of publicity and a productive one in terms of work: the mesh will become stable, sugar will become more predictable, and we will see some rocking learning activities emerge. Deployment teams around the world will take a cool, sexy prototype device and find the practical, unsexy methods to make it meet the needs of kids and teachers in the developing world.
For us in Nepal, 2008 is the year of we hope to get the implementation right. We hope to create an implementation model that works and do a proper evaluation of this project to prove to the naysayers what we ourselves know to be true: XO's + Constructionist Education can really transform how kids (and adults) learn according qualitative and quantitative measures. Our pilot is tiny compared to those in Peru, Uruguay, and Mexico but we believe that we are laying the groundwork for implementing this project across Nepal.
I'll get off my soapbox now and get some dinner. Mmm, buffalo choila and chiura.