of is a primitive that allows us to reach into an agent and pull out the value of a variable it owns. Normally, agent variables (such as color, size, shape, as well as variables defined with <agents>-own commands) are accessed within an agent context using the ask primitive. of allows us to get to those variables outside of such a block or to get all the values for each agent in an agentset all at once as a list. For example, if we wanted to create model where each person picked a random friend and gave them an apple if they didn't already have one, we would write the following code:
ask people [
let my-friend one-of other people
if [apple] of my-friend = 0 [
ask my-friend [ set apple apple + 1 ]
]
]
Things to keep in mind when using of:
of is very useful to query agent characteristics such as finding the average size of turtles (mean [size] of turtles), the largest turtle size (max [size] of turtles), etc.of is also very useful in creating monitors or plots in the interface. Oftentimes, we will need to present some sort of a summary value such as the average wealth of turtles (mean [wealth] of turtles) or total wealth of turtles (sum [wealth] of turtles).of such as [checking-account + savings-account] of customers to get the total money in customers' accounts or [size / 2] of circles to automatically calculate the radius of circles in a model.of with agentsets (and without primitives such as mean), it will report a list. For example, if we have 3 turtles in a model, [size] of turtles would report [1 1 1].In the model example below, we have a model that is similar to the popular online game agar.io. We have many turtles that represent circles and they move around randomly. When two turtles touch each other, the larger turtle eats the small turtle. We use of to compare the two touching turtles within an ask statement in order to pick the larger one and then to add the sizes of the two touching turtles.