Monday and Tuesday of this week, I had the priviledge of meeting and talking to Joe Polastre, CTO of Sentilla formerly MoteIV. My thesis advisor Philippe Bonnet had invited Joe to Denmark. Monday Joe gave a talk about Sentillas java based sensor network platform, spized up with a fun sensor network case study they had conducted during JavaOne.
Apparently they had mounted an array of sensors in the doorways of the large auditoriums at Moscone Center, telling the direction people where moving in, so when people were entering and leaving the room, they kept count. It gave some interesting metrics - as to which speakers people REALLY enjoyed. Fx. while people where watching the keynote with Rich Green and waiting for Neil Young, practically no one was leaving the room. This was a fine example of pervasive computing, because people weren't aware of the the sensors.
Tuesday I had the opportunity to sit down with Joe for an hour and talk to him about my thesis. So what's it about people keep asking me - well - we all know physics experiments, chemistry experiments and in general science experiments from high school. Yet teaching computer science is very abstract - so would it be possible to teach computer science at high school level by using experiments conducted on the SunSPOT? And if yes, what kind of experiments could be conducted, that are both fun and fullfill their computer science curriculum?
My wish is to write a small interpreted language, that can run on the SunSPOTs within the Squawk VM, in which you can write small programs - and the programs are the way to "solve" the experiments. Hopefully I'll have something a little more substantial than just the thoughts within the next couple of months.
Anyway Sentillas sensor network platform was really cool, because of its scale - it even won the prestiguos Duke Choice Award at JavaOne. Thanks to Joe for taking his time to visit us in little Denmark 
Posted by Sentilla on July 09, 2008 at 08:36 AM CEST #