08 July 2007
I knew this day would come, but I never would have guessed it would come so quickly. I guess the old saying is true. Time really flies when you’re having fun.
That’s right folks. I’ve finished my studies and passed my exams. I’m officially a college graduate! (Real world beware…)
You may hear yourself asking: “Wait a second, Joe graduated a month ago! Why is he posting now? He must be really lazy.”
Better Late than Never.
Well you’re wrong! I am lazy, but that’s not why I’m graduating now. As part of my final studies I was fortunate enough to be involved in multidisciplinary project run through the Information Technology department under a course named “Innovation and Invention”. Because I am a Computer Science graduate, and because my successful graduation hinged on completing a final computer science elective, I was forced to enroll in the aforementioned totally awesome class as a “computer science independent study”.
How did I end up in this situation? Why would I bother to go through so much trouble for one class?
It was the coolest course I’ve ever enrolled in!
During my second co-op stint at Paetec Communications I was tipped off to a convention/speaker series by the name of TED (Technology Entertainment and Design). These talks cover a broad spectrum of topics from cool new user interfaces to educational beliefs and systems. As a part of one of these talks, I stumbled across the new (well not so new, last year new) multi-touch display presented by Jeff Han. Now, as Jeff mentions in his talk, multi-touch technology is not new, however, by utilizing a physical concept called Frustrated Total Internal Refraction Jeff has devised a way to make multi-touch displays of varying sizes while staying relatively affordable. I thought this was the coolest thing since Velcro, but that was that.
Fast forward a few months and my room mate Jason Morrison noticed that there’s an entire community based around development and experimentation with multi-touch displays utilizing Jeff Han’s method! On top of that, there’s an Open Source library by the name of TouchLib which encapsulates functionality from Intel’s Open Computer Vision into a form that developers for multi-touch displays can use to create cool applications. Jay was hooked, and decided to build one. I literally followed Jay around every step of the way (arguably being one of the worst kind of nuisances) desperately looking for some way to become involved in his new pet project.
We can get course credit for this?!
Shortly after ordering the parts, Jason was clued into this Innovation and Invention course. After speaking with the professor, it was decided that the multi-touch display that Jason was working on might be a good candidate for a side project within the Innovation and Inventions course load. Sure enough, work could be done on the multi-touch display and course credit could be received for it.
After speaking with Jason to make sure I wasn’t treading too far on his domain, I contacted Dr. Jon Schull about the possibility of working with the course in some capacity. I was invited to attend the first course and see what I thought. Each course was set up in a seminar/workshop (read: long hours) which gave us one long scheduled block at the end of each week in which to officially meet. After a rousing speech by our all powerful leader Jon I was hooked. The course seemed like a great opportunity to complete my graduation requirements while working on a project like nothing I had ever seen before. I dropped another class which had not been keeping my interest and scheduled my new independent study.
What the heck is Innovation and Invention?
At this point it is worth noting that the multi-touch display was not the focus of the Innovation and Inventions course. The true focus of the course was what came to be known as the “Immersive Dome” project.
The “Immersive Dome” project involved:
When the above pieces were mixed vigorously together they were to form a geodesic dome which, when rear projected uponm would create an interactive 360 degree immersive environment. One thing you need to keep in mind when creating an interactive immersive environment is how, exactly, you’re going to let people interact with th environment!
This is where I come in.
During the first class each student introduced themselves, their background, and their interests. Once we had compiled a list of applicable skill sets (almost everything was applicable, from RPC programming experience, all the way to sewing skills) we were able to place each member of the course into at least one project group. These smaller groups would then meet on their own to achieve their goals for the week to help move the dome project forward.
I had described myself as a Computer Science major with experience programming with interests in computer vision and multi-touch interface design. As such I was put into the “software infrastructure”, “blob-detection”, and “multi-touch” groups. Each group was related, but had a different specific goal in mind.
The software infrastructure group was tasked with developing and coordinating a set of software applications which would coordinate and control the launching and interaction with the many different software packages which would be running on the dome.
The blob-detection group was tasked with discovering the best methods by which to utilize computer vision to track one of many things involving blobs. In some cases we were tracking human beings (positions within the dome for movement controlled by humans walking), in other cases we were tracking finger presses (multi-touch interface design), and in either case we had our hands full.
The multi-touch group was focused on getting the multi-touch interface up and running in such a way that it worked, and our multi-disciplinary group could develop applications which would utilize it’s unique multi-touch design. We ended up settling on using TouchLib on a dedicated computer vision box packaging the touch data up as OSC [pdf] packets which would then be sent to one of many services which could utilize the touch data.
Jason Morrison, Matt Kampschmidt, and myself, as representatives of the multi-touch display group had the distinct honor of both getting our system up and running within the first three weeks of class, as well as being the first group to complete their objectives all together.
Success and Celebration.
Our excitement was palpable. Unfortunately the multi-touch display as an interface for the 360 degree dome was eventually scrapped in favor of directly tracking humans themselves as they walk within the dome. We were, however, able to get a multi-touch surface up and running, design software for the surface, and demonstrate that both worked together flawlessly. Combine that project work with a relatively long paper describing my development efforts on both the dome, and the multi-touch display, and I had fulfilled my independent study requirement.
The dome project, on the other hand, was a complete success. The dome was displayed as a part of the Creativity and Invention Symposium at the RIT. We entertained multiple visitors with a 3d fly-through of a monkey brain, 360 degree immersive Google Earth experiment, and a visitor controlled walk through of a digital story book world.
After the project completed, the dome was displayed for a short while at BarCamp Rochester 2. Jon Schull and Lindsay Cade (an amazing artist, technical writer, engineer, and seamstress all rolled into one), Dean Ganskop, and Adam Weill also presented on the dome at the New Media Consortium Summer 2007 Conference.
And thus my graduation was delayed.
Unfortunately for me, the conferences and presentations listed above lasted for quite a long time, during which our professor was unable to update grade records. Once he returned, and received my many (in many case overly frantic) e-mails wondering why my grade had not been applied, he awarded my grade and I officially graduated.
This course by far was one of the most rewarding, and interesting experiences of my college career. Everybody would do well to participate in some sort of large scale multidisciplinary project. Hopefully, if Jon has his way, every student will.