I had an interesting conversation today with Bret Victor, which touched on the question of how much constructionism there should be in an interactive presentation. In other words, how much should you be able to build your presentation of a simulation out of general components, right then and there in the class-room?
The distinction touches on the difference between “live” and “canned” presentations. There are people who give spectacular talks, with truly exciting visuals. Yet during the question and answer session those speakers fall back on words. All of the spectacular visuals were pre-cooked before the show.
Bret has a vision, which I admire, that in an ideal world any question can be answered using the component pieces — right then and there — of the very tool you used to give the talk. It’s an ideal of a procedural visual language.
I find that I fall somewhere in the middle between Bret’s vision and the more traditional approach. I have a practical need to lecture on certain subjects, like computer graphics. I know that if I focus all my energies on building first-principles tools for giving those lectures, that may be all I ever have time to do.
And yet I agree with Bret that it is better to answer questions from the class using the procedural tools of my presentation, if I can.
In the end, we will all play our part. I may focus my energies more teaching computer graphics, and Bret may focus more on presentation tool as real-time visually authorable simulation, but I do believe that we, and others working in this space, will eventually converge on a new kind of literacy.