[jerf]

The optimal teaching approach may depend on the level of the material, too. Slides & notes mailed to students may be a great option for "routine" math courses, things taken by a ton of different majors, where the paths are ground so smooth by the millions of feet traveling it that the entire class can be called in advance by even a fairly green instructor, almost down to the questions they're going to get.

As you get into higher math, though, the chalkboard becomes more important, because it's a lot more like working it out in front of the student. The class is far more likely to ask questions that are going to consume a couple of boards of scribbles to answer, and the professor may not be able to anticipate it all in advance. Trying to do this on a slide is very inconvenient and I'm still yet to see a computer interface in common use that can pull it off at all.

Many of those classes I took were actually hybrid; there was a core set of projected slides that could be mailed out (though they weren't always) that set the agenda and gave the basics and results, but there was a lot of chalkboard use as well, since as long as the room is even modestly large, there's no particular need to use just one or the other.

Graduate-level graph theory in particular used the chalkboard pretty heavily for what were essentially impromptu animations, between the numbers being added to a diagram as whatever algorithm we were learning about progressed, and the gestures being used by the professor.

[Aromasin]

That certainly makes much more sense. My experience was primarily undergraduate, plus a years masters. I will admit, during the MSc chalk-board work was valuable during the seminars and labs, where it was much more design based.