Monday, May 14, 2007

Ondelettes are adorable.

W00t. Just sent in a paper on using swarm robotics as a model for classrooms (yes, this is related to Boris's project). It's the counterpart to his model of the teacher-student broadcast system, which works well for lectures; my premise was that student-driven small group projects are better modeled by a purely peer-to-peer system of mobile robots with a mesh network (no teacher-arbiter). Along the way, I learned about wavelets* and now want to learn more; apparently Gilbert Strang, one of my textbook-heroes, has written a book on it... and the book is expensive. Plus I have many more books to read. And I can always hang out in MIT's libraries this summer, and I know they have that book because I looked it up for a friend there years ago. Yes. That's a better plan.



What's left: a SCOPE poster (which is waiting on Eric Gallimore), editing a paper on the history of Olin's curriculum revisions, and then a Thursday afternoon presentation for my anthropology class which is half-meta; the first part is on observations of an engineering student (me) learning anthropology and why it's been both tremendously difficult and incredibly enlightening, the second part on my research proposal to study the subcultures of engineering education in universities around the world. I was originally focusing solely on pedagogical techniques, but conversations with Pres. Miller and feedback from the discussions at the President's Council meeting have persuaded me that it's the pervasive culture of a place that makes a difference in student learning more than a mere tally of what methods are used in the classrooms. (Culture, of course, is much harder to "pin down." If I'm not careful I might end up with a doctoral thesis on my hands.**) More about this later.



The sun rises. Time to sleep. I'm exhausted and a weird mix of conflicting emotions right now, but the dominant feeling is peaceful happiness so I'm going to run with that and just fall into bed for a couple of hours.



* in French, ondelettes. Note: Naming your ideas well is important. Half the fun of learning about wavelets is being able to say a word that means "little waves" over and over again; it's the signal processing equivalent of calling it the iNoun. Instant theoretical coolness.

**Not...that I'd mind that, really.

1 comment:

Herbert Chang said...

ondelettes are cute!! hehe I just did a sigsys project-y thingie looking into ondelettes. Now Ithink all dsp is liek wavelets though. no more of that silly Fourier stuff =P