The YouTube clips I show in my distributed systems class
I made it a tradition to start my CSE 4/586: Distributed Systems lectures with a short YouTube video. As the students trickle into their seats, I play a YouTube video related to the lecture of the day.
Video is an underutilized medium in teaching. I don't understand why more professors do not use them. They are visual and they wake the students up, and get them interested in the class. After the students are primed with the video clip about the context of the lecture, they listen the class more attentively. I get overwhelmingly positive feedback from the students over many years about this practice.
Here are some of the videos I show in my class.
This is the video I show in my first class when I introduce the definition of a distributed system. I tell them a murmuration of starlings can be considered a distributed system, because the definition fits: A collection of autonomous nodes communicating to address a problem collectively, with no shared memory, no common physical clock, and geographical separation. Come to think of it, maybe this game of life video is more relevant, because it clearly lays out the 4 simple local rules that create the amazing emergent behavior.
In the first couple weeks of the class I teach about specifying and reasoning about the safety and progress properties of distributed programs. Some students find this short forage into predicate logic and computation model tedious. So to motivate them I tell them that this understanding of program properties and reasoning about them will get very handy as they learn more complicated algorithms soon. I tell them to trust me on this: /before they can do karate, they need to learn about some tedious wax on and wax off/. Then I show them scenes from Karate Kid, an awesome movie from my childhood. The movie is from 1984, and recently I came to the sobering realization that it is older than my students in the class.
When I teach about the distributed system computational model, I show a short clip from a lecture by Leslie Lamport, a forefather of distributed systems.
When I teach about examples on reasoning about distributed programs, I show them inevitably some unrelated clips. But I believe these clips are extremely helpful to familiarize the students with the CSE field. Some of my favorites are: Jeff Bezos (CEO Amazon) talks about Internet entrepreneurship and about Xerox PARC and the origins of the PC.
When I introduce logical time in distributed systems, of course, I show a clip from Genius: The life of Albert Einstein.
When I talk to them about mutual exclusion, I show them a video of toy train track mutual exclusion. I follow that up with a traffic intersection mutual exclusion. And when I talk about dining philosophers, I show a simulation of dining philosophers.
For consensus, I don't have anything good to show. Instead I bring a couple students to the front of the class and we re-enact the attacking generals problem with messengers running between the two generals.
For chain replication, I show a video about DNA replication. Maybe unrelated, but super interesting.
For failure detectors, I use this skit from Monty Python. This may be the most appropriate use of a Monty Python skit in a distributed systems class.
When I teach them about fault tolerance, I show them this clip about resilience. When I tell them about the concept of masking of fault-tolerance and robust-yet-fragile concept, I show them about this video. In this earlier post, I explain the concept and the relevance of this clip.
I have been using some of these videos for 4-5 years now. So, if you have suggestions for alternatives, I will be happy to upgrade my YouTube video curation.
Video is an underutilized medium in teaching. I don't understand why more professors do not use them. They are visual and they wake the students up, and get them interested in the class. After the students are primed with the video clip about the context of the lecture, they listen the class more attentively. I get overwhelmingly positive feedback from the students over many years about this practice.
Here are some of the videos I show in my class.
This is the video I show in my first class when I introduce the definition of a distributed system. I tell them a murmuration of starlings can be considered a distributed system, because the definition fits: A collection of autonomous nodes communicating to address a problem collectively, with no shared memory, no common physical clock, and geographical separation. Come to think of it, maybe this game of life video is more relevant, because it clearly lays out the 4 simple local rules that create the amazing emergent behavior.
In the first couple weeks of the class I teach about specifying and reasoning about the safety and progress properties of distributed programs. Some students find this short forage into predicate logic and computation model tedious. So to motivate them I tell them that this understanding of program properties and reasoning about them will get very handy as they learn more complicated algorithms soon. I tell them to trust me on this: /before they can do karate, they need to learn about some tedious wax on and wax off/. Then I show them scenes from Karate Kid, an awesome movie from my childhood. The movie is from 1984, and recently I came to the sobering realization that it is older than my students in the class.
When I teach about the distributed system computational model, I show a short clip from a lecture by Leslie Lamport, a forefather of distributed systems.
When I teach about examples on reasoning about distributed programs, I show them inevitably some unrelated clips. But I believe these clips are extremely helpful to familiarize the students with the CSE field. Some of my favorites are: Jeff Bezos (CEO Amazon) talks about Internet entrepreneurship and about Xerox PARC and the origins of the PC.
When I introduce logical time in distributed systems, of course, I show a clip from Genius: The life of Albert Einstein.
When I talk to them about mutual exclusion, I show them a video of toy train track mutual exclusion. I follow that up with a traffic intersection mutual exclusion. And when I talk about dining philosophers, I show a simulation of dining philosophers.
For consensus, I don't have anything good to show. Instead I bring a couple students to the front of the class and we re-enact the attacking generals problem with messengers running between the two generals.
For chain replication, I show a video about DNA replication. Maybe unrelated, but super interesting.
For failure detectors, I use this skit from Monty Python. This may be the most appropriate use of a Monty Python skit in a distributed systems class.
When I teach them about fault tolerance, I show them this clip about resilience. When I tell them about the concept of masking of fault-tolerance and robust-yet-fragile concept, I show them about this video. In this earlier post, I explain the concept and the relevance of this clip.
I have been using some of these videos for 4-5 years now. So, if you have suggestions for alternatives, I will be happy to upgrade my YouTube video curation.
Comments