Recording lectures? Posting the Recordings? Using Slides?

The issue of whether to record lectures or post slides or more generally how much material to give to the students is a new question (the last 20 years?) but I do have an anecdote from 1978.

I was taking an applied math course on Mathematical Modelling from James C Frauenthal (He would sometimes write his initials \(\binom{J}{F}\)) and he passed out his notes ahead of time. I think I was the only student who read them ahead of time. One time I had read the following passage ahead of time:

We have been somewhat cavalier in our assumptions.

During class he said

What is wrong with this mathematical model? 

I replied

We have been somewhat cavalier in our assumptions.

He was somewhat surprised, but pleased that someone was reading his notes. 

FAST FORWARD TO MODERN DAY.

How much material should we make available for our students? I post slides and recordings. 

PRO: If a student misses class they can catch up. Especially good if missing class is rare.

PRO: If a student is in class then they can go back to some point they were confused on.  

PRO for slides: When asking a student when they began getting lost we can find the exact slide. This is much better than the word salad that students sometimes emit when describing where they are lost.

BILL: So you understood the definition of P. So you were lost when I defined NP? 

STUDENT: No, I got lost when you described some kind of really exciting algorithm.

BILL: Exciting algorithm? What did it do?

STUDENT: You said it was a paradox.

BILL: This is a class in algorithms. We have not discussed any paradoxes.

STUDENT: Did so!

BILL: We can figure out what ails thee. What did the algorithm do?

STUDENT: Something about the whole being greater than the sum of its parts.

BILL: Parts! I think you mean that we solve sub parts and then put them together. This is the Dynamic Programming paradigm. OH- I think you confused  paradigm and paradox.

STUDENT:  That's exactly what I said. Dynamic means exciting! And paradox is just another name for paradigm.

Often it was hard to see where they got lost.  

CON: Students may skip class and not go over the slides or recordings!

CON: The technology sometimes does not work.

BILL: You missed class and expect me to redo the lecture in my office. Did you watch the recording?

STUDENT: No. The recording did not work and its your fault!

BILL:  The first day of class I said you should come to class for the following reasons

1) You can ask questions. The paradox is that's hard to do in a large class, but with so many student cutting class, its a small class!

2)  Taking notes is a good way to solidify knowledge.

3) Going to class forces you to keep up.

4) The technology might not work. Last semester this happened four times. Twice it was my fault, and twice is was not my fault. But that does not matter- it will happen. 

5) If  you show up in my office hours and want me to explain  what I lectured on I will be annoyed.

STUDENT: Uh,... I missed the first day.

CON: In the long term students get in the habit of not going to class.  I can't tell if this is worse than it used to be. 

CON for Slides: Its hard to be spontaneous. Some of the classrooms don't even have whiteboards to go off-script with. The following happened in the pre-slide days (apologies- I've told this story before on this  blog) on April 25, 2003 in my Automata Theory class. I had already done decidability and was goin going to do r.e. sets.

STUDENT: Do you know whose 100th birthday it is today?

BILL: Will there be cake? If so will they let me eat cake?

STUDENT: Today is Kolmogorov's 100th birthday.

BILL: AH! I was going to do r.e. sets but instead I will do Kolmogorov Complexity!

STUDENT: Great! Uh. Maybe. Is it hard? 

BILL: Its fun!

I then gave a lecture on Kolmogorov complexity on the fly, on the whiteboard. I made it part of the course and on the final I asked them to show that if  w is a K-random string of length n then any context free grammar for {w} in Chomsky Normal Form requires at least \( n^{0.99} \) rules (this is not the strongest result possible). 

This is impossible with slides. No more on-the-fly lectures. 

CON for slides: Some proofs are impossible to do really do on slides. The Poly VDW theorem and the Kruskal Tree Theorem are two of them. Fortunately those are both in Ramsey Theory that has 30 students and a whiteboard, so for those lectures I use a white board. 

PRO for slides: My handwriting isn't that good, so slides helps a great deal.

CAVEAT: I used to read a proof, write it out by hand, type it up in LaTeX, and then make  slides.  Now I go straight from reading it to slides. This is sometimes not a good idea as I am worrying about fonts and  formatting before I really understand the proof. I recently went BACK to the handwritten notes  THEN LaTeX THEN slides. That increased my understanding since (1) when doing the handwritten notes I was not distracted by fonts or formatting, and (2) at every iteration I picked up some subtle point I had missed. 

CAVEAT: When teaching a large class you really can't use the whiteboard since the people in the people in the back row can't see. I don't know if that's an argument FOR slides or AGAINST large classes. 

SO- what do you do: record, not record, slides, no slides.And why? And does it work?