The class started with a counting game as an icebreaker. As a class, the students were required to count by 1 as high as they could go. However, if the number was a multiple of 3 or had the number 3 in it, they were supposed to clap and not say the number. This proved to be a challenging yet seemingly fun exercise for the class.
Next the class started going over the concept of why we would use a model. The cookies that were meant for the workshop students were stolen and a good way to track down stolen items is to have a dog sniff them out! The students began building a model of this because no one had an actual hound with them of course. There were 3 variables, the hound, the cookies, and the background (the office). With the variables established, the class started a new AgentSheets model. Once a new sheet was made, the class imported the photos of the variables as the agents. Everyone made a simple function where the dog moved randomly. Then they added a barking noise to the move-random function. They were given the opportunity to experiment with the model and find the best way to accomplish the tasks. Another challenge for the class was to assign a key on the keyboard that would trigger the barking noise. The goal of the model was to program the dog to sniff the office and find the cookies. Eventually, everyone was able to assign a value to the cookies; the higher the value, the stronger the smell of the cookie is.
By the end of the class, almost everyone had a functional model in which a dog moved randomly around and once he smelled the cookie he would move towards it. This modeled exactly how a dog would act in order to find drugs, missing people, or in this case, the missing cookies!