Met with Martin Brooks this afternoon
I had another crazy meeting with Martin Brooks. Martin is the closest thing to a mad scientist I’ve ever seen. But he is a friendly mad scientist. Right now, he is working with a Toronto company for the Japan World Fair. I could not google this event? Martin is convinced that broadband will change the world. He recently completed that Music Grid project which allowed people from all over the world to teach each other about music, including kids in very remote locations where they’d never could have had any music lessons.
They are setting up a very fancy videoconferencing station with motion sensors and all, in UQAM’s Président-Kennedy building (near Place-des-Arts). As far as I can tell you can just go there and stop by and look at the gear. It is pretty amazing, I was there this afternoon. They are connected live with France, to some museum over there. You can go on a stand and the system will not only record your image and voice, but you can point your finger and it will detect the motion and you can use your hands to… well, I couldn’t quite get the full picture… but it looks amazing. This seems related to what some UQAM people do in arts with kids this summer. Sorry, didn’t get all the details. But quite possibly, some like Montreal kid will get to play with some broadband equipment I can only dream about.
Actually Martin wanted to tell me about his latest work in monotonicity preserving simplification and not about broadband. He came up with the following statement: if you interpolate any real data in such a way that you don’t add any extrema, you get a function which has way too much complexity for any sensible modelling. He gets to this result by interpolating data such as images, and then he computes contour lines using some crazy lisp code on a very old Mac laptop (did I mention he was like a mad scientist?) and he gets extremely complex contour lines… much more so than you’d expect. Interesting.
