Friday, May 9, 2008

Randy Pausch's Last Lecture Part-8

It was a great teamwork. I think it was a great yin and a yang, but it was more like YIN and yang. And he deserves that credit and I give it to him because the ETC is a wonderful place. And he's now running it and he's taking it global. We'll talk about that in a second.

Describing the ETC is really hard, and I finally found a metaphor. Telling people about the ETC is like describing Cirque du Soleil if they've never seen it. Sooner or later you're going to make the mistake. You're going to say, well it's like a circus. And then you're dragged into this conversation about oh, how many tigers, how many lions, how many trapeze acts? And that misses the whole point. So when we say we're a master's degree, we're really not like any master's degree you've ever seen.

Here's the curriculum [Shows slide of ETC curriculum, listing "Project Course" as the only course each semester; audience laughs] The curriculum ended up looking like this. [shows slightly more detailed slide]. All I want to do is visually communicate to you that you do five projects in Building Virtual Worlds, then you do three more. All of your time is spent in small teams making stuff. None of that book learning thing. Don and I had no patience for the book learning thing. It's a master's degree. They already spent four years doing book learning. By now they should have read all the books.

The keys to success were that Carnegie Mellon gave us the reins. Completely gave us the reins. We had no deans to report to. We reported directly to the provost, which is great because the provost is way too busy to watch you carefully. [laughter] We were given explicit license to break the mold.

It was all project based. It was intense, it was fun, and we took field trips! Every spring semester in January, we took all 50 students in the first year class and we'd take them out to Pixar, Industrial Light and Magic, and of course when you've got guys like Tommy there acting as host, right, it's pretty easy to get entrée to these places. So we did things very, very differently. The kind of projects students would do, we did a lot of what we'd call edutainment.

We developed a bunch of things with the Fire Department of New York, a network simulator for training firefighters, using video game-ish type technology to teach people useful things. That's not bad. Companies did this strange thing. They put in writing, we promise to hire your students. I've got the EA and Activision ones here. I think there are now, how many, five? Drew knows I bet.

[Drew Davison, head of ETC-Pittsburgh, gestures with five fingers]. So there are five written agreements. I don't know of any other school that has this kind of written agreement with any company. And so that's a real statement. And these are multiple year things, so they're agreeing to hire people for summer internships that we have not admitted yet. That's a pretty strong statement about the quality of the program.

And Don, as I said, he's now, he's crazy. In a wonderful complimentary way. He's doing these things where I'm like, oh my god. He's not here tonight because he's in Singapore because there's going to be an ETC campus in Singapore. There's already one in Australia and there's going to be on in Korea. So this is becoming a global phenomenon. So I think this really speaks volumes about all the other universities. It's really true that Carnegie Mellon is the only university that can do this. We just have to do it all over the world now.

One other big success about the ETC is teaching people about feedback [puts up bar chart where students are (anonymous) listed on a scale labeled "how easy to work with" ] -- oh I hear the nervous laughter from the students. I had forgotten the delayed shock therapy effect of these bar charts. When you're taking Building Virtual Worlds, every two weeks we get peer feedback. We put that all into a big spreadsheet and at the end of the semester, you had three teammates per project, five projects, that's 15 data points, that's statistically valid.

And you get a bar chart telling you on a ranking of how easy you are to work with, where you stacked up against your peers. Boy that's hard feedback to ignore. Some still managed. [laughter] But for the most part, people looked at that and went, wow, I've got to take it up a notch. I better start thinking about what I'm saying to people in these meetings. And that is the best gift an educator can give is to get somebody to become self reflective.

So the ETC was wonderful, but even the ETC and even as Don scales it around the globe, it's still very labor intensive, you know. It's not Tommy one-at-a-time. It's not a research group ten at a time. It's 50 or 100 at a time per campus times four campuses. But I wanted something infinitely scalable. Scalable to the point where millions or tens of millions of people could chase their dreams with something. And you know, I guess that kind of a goal really does make me the Mad Hatter. [Puts on a Mad Hatter's green top hat].

So Alice is a project that we worked on for a long, long time. It's a novel way to teach computer programming. Kids make movies and games. The head fake -- again, we're back to the head fakes. The best way to teach somebody something is to have them think they're learning something else. I've done it my whole career. And the head fake here is that they're learning to program but they just think they're making movies and video games.

This thing has already been downloaded well over a million times. There are eight textbooks that have been written about it. Ten percent of U.S. colleges are using it now. And it's not the good stuff yet. The good stuff is coming in the next version. I, like Moses, get to see the promised land, but I won't get to set foot in it. And that's OK, because I can see it. And the vision is clear. Millions of kids having fun while learning something hard. That's pretty cool. I can deal with that as a legacy.


No comments: