Prof. Richard Lipton (of GeorgiaTech) recently mused aloud on his blog about the (over)use of slides in talks these days. I left a comment on his blog yesterday, and then realized that my reply could make for a separate post all by itself. So, let me note here what I wrote.
Prof. Lipton’s posts, it seems, typically end with a section of the title: “Open Problems.” These are not some open technical problems of research, but just some questions of general interest that he raises for discussions. My reply was in reference to such a question:
“Do you like PowerPoint type talks or chalk talks?”
In the next section I give my (slightly edited) reply.
* * * * * * * * * * * * * * *
Both. (I am talking about the Open Problem.) Simultaneously. And, actually, something more.
On the rare occasions that Indians have allowed me to teach (I mean in the recent past of the last decade or so), I have found that the best strategy is:
(i) to use slides on a side-screen (and, actually, the plain wall besides the white-/black-board works great!)
(ii) and to keep the white-/black-board at the center, and extensively use it to explain the points as they systematically appear on the slides.
It’s the same old bones-and-flesh story, really speaking. Both are necessary.
Even if you don’t use the slides for a lecture, preparing them is still advisable, because a lot of forethought thereby has to go into structuring your presentation, including some thought about the amount of time to be allocated to each sub-unit of a lecture. Why, if you prepare the slides, even if you don’t use them, you would find that your management of the black-board space also has improved a great deal!
I know quite a few professors who seem to take the ability to deliver a lecture without referring to any notes or slides, as the standard by which to judge the lecturer’s own mastery of the subject matter. Quite fallacious, but somehow, at least in India—where there always is a great emphasis on memorization rather than on understanding—this opinion persists widely. … And, that way, the lecturer’s own mastery, even if necessary, is not at all sufficient to generate a great lecture, anyway. Newton in Cambridge would teach mostly to empty benches.
The flow of a spontaneously delivered lecture is fine, as far as it goes. But far too many professors have this tendency to take things—students, actually!—for granted. If you don’t prepare slides, it’s easy—far too easy, in fact—to do that.
I have known the “free-flowing” type of professors who would habitually dwell on the initial and actually simpler part of a lecture for far too long (e.g., spend time in drawing the reference diagrams (think FBD of mechanics) or in discussing the various simple parts of a simple definition, etc.), and then realizing that “portion is to be finished,” hurriedly wind up the really important points in the last five minutes or so. For example, I have seen some idiots spend up to 45 minutes explaining the simplest points such as, say, the household plumbing as an analogy for the electric circuits, or the movement of a worm as an analogy for the edge-dislocation motion, and then wind up in the remaining 15 minutes the really important topics like the Thevenin/Norton network theorem (EE) or mechanisms of dislocation growth (materials science).
Maths types are the worst as far as the so highly prized “flow”—and showing one’s genius by solving problems on the fly on the black-board without ever referring to notes—goes. Well, if you are going to prize your own genius in that manner, some other things are going to get sacrificed. And, indeed, they do! Ask yourself if your UG ODE/PDE teacher had appropriately emphasized the practically important points such as the well posed-ness of the DEs, or, for that matter, even the order of a DE and the number of auxiliary (boundary and/or initial) conditions that must be specified, and how—whether you can have both the flux and the field conditions specified at the same point or not, and why. I have known people who got these points only while taking post-graduate engineering courses like CFD or FEM, but not from the UG mathematics professors proper. The latter were busy being geniuses—i.e. calculating, without referring to notes or slides.
All such folks must face the arrogance of their idiocy if (to them) the highly constraining rule that slides must be prepared in advance for every lecture, is made compulsory. It should be!
You can never make anything fool-proof, of course. For every “free-flowing” guy of the above kind, there always is that despicable “nerdy” type… You know, one who meekly slides into his class with his slides/notes, hangs in nervously there for the duration of the lecture, puts up the slides and reads them aloud (if you are lucky, that is—I have suffered through some specimens who would merely mumble while vaguely looking somewhere in the direction of the slides), “solves” the problems already solved in the prescribed text-book, and then after a while, leaves the class with somewhat less meekness: carrying on his face some apparent sense of some satisfaction—of what kind, he alone knows. These types could certainly do well with the advise to do some chalk-work.
With slides, diagrams can be far more neat; students can copy definitions/points at their own convenience during the lecture, and you don’t have to wait for every one in the class to finish taking down the material before proceeding further because they know that hand-outs would be available anyway (because these are very easy to generate); and the best part: you don’t have to worry too much about your hand-writing.
In between PowerPoint and LaTeX, I personally like Beamer because: (i) its template makes me feel guilty any time I exceed three main points for an hour-long lecture (though I often do!), and (ii) I always have the assurance that the fonts won’t accidentally change, or that the diagrams wouldn’t begin floating with those dashed rectangles right in the middle of a lecture. And, it is free. To a mostly jobless guy like me, that helps.
Finally, one word about “more.” Apart from chalkwork and slideshows, there are many other modalities. Even simplest physical experimentation is often very useful (e.g., tearing a paper in a class on fracture mechanics, or explaining graph algorithms using knotted strings, say as hung down from this knot vs that knot, etc). Physical experimentation also kills the monotony and the boredom.
In my class, I also try to use simulations as much as possible. By simulation/animation, I do not mean the irritating things like those alphabets coming dancing down on a PowerPoint slide as if some reverse kind of a virus had hit the computer. I mean real simulations. For instance, in teaching solid mechanics, FEM simulation of stress fields is greatly useful. I gained some of the most valuable insights into classical EM only by watching animations, e.g., noticing how changing an electric current changes the magnetic field everywhere “simultaneously” (i.e. at each time step, even if the propagation of the disturbance is limited by ‘c’). In CS, you can spend one whole hour throwing slides on the screen, or doing chalk-work, or even stepping through code to explain how, e.g., the quicksort differs from the bubble sort, but a simple graphical visualization/animation showing the sorting process in action, delievers a certain key point within 5 minutes flat; it also concretizes the understanding in a way that would be imposible to achieve using any other means.
My two/three cents.
* * * * * * * * * * * * * * *
See Prof. Lipton’s post and the other replies he got, here [^]; my original reply is here [^]. BTW, as an aside, when I wrote in my reply at his blog, the two replies immediately above mine (i.e., those by “CrackPot” and Prof. M. Vidyasagar) had not yet appeared, and so, there is no implied reference to them, or, for that matter, to any replies earlier either—-I just wrote whatever I did, in reference to the main post and the “Open Problems” question.
* * * * * * * * * * * * * * *
A Song I Like:
The “chalk-work” version:
(French) “L’amour est bleu”
Music: Andre Popp
Singer: Vicky Leandros
Lyrics: Pierre Cour
The “slide-show” version:
(Western instrumental): “Love is blue”
Orchestrator and Conductor: Paul Mauriat
[Asides: I had first heard Mauriat’s version on a cassette that I had generally bought sometime in the late 1980s, and till date was not at all aware that there also was an actual song with actual lyrics, here. I always thought that it was some instrumental composition by someone, all by itself. I saw the video of Vicky Leondros’ version only today, after an Internet search. And, the browsing (mostly Wiki!) also reveals that the singer’s real name was Vassiliki Papathanassiou, not Vicky Leandros. Though in the video recording she sometimes looks Asian, she actually was a Greek singer who sang in French while representing Luxemborg in the 1967 Eurovision Song competition, where she was placed 4th. And, Mauriat’s version, per Wiki, “became the only number-one hit by a French artist to top [sic] the “Billboard Hot 100” in America.“ … Tch… Are they sure about this last bit? I mean, it serves to give just too much credit to the Americans, don’t you think?]