"DAG (directed, acyclic graph). One common use of graphs is to represent flows of dependencies, for instance the prerequisites that different college courses have for one another. Another example is project management workflows: the tasks needed to complete a project become vertices, and then the dependencies they have on one another become edges. The graph in Figure 5.4 shows the steps in making a batch of brownies, and how these steps depend on each other. The eggs have to be cracked before the ingredients can be mixed, and the oven has to be preheated before baking, but the pan can be greased any old time, provided that it’s done before pouring the brown goop into it.\n\nA graph of dependencies like this must be both directed and acyclic, or it wouldn’t make sense. Directed, of course, means that task X can require task Y to be completed before it, without the reverse also being true. If they both depended on each other, we’d have an infinite loop, and no brownies could ever get baked! Acyclic means that no kind of cycle can exist in the graph, even one that goes through multiple vertices. Such a cycle would again result in an infinite loop, making the project hopeless. Imagine if there were an arrow from bake for 30 mins back to grease pan in Figure 5.4. Then, we’d have to grease the pan before pouring the goop into it, and we’d have to pour the goop before baking, but we’d also have to bake before greasing the pan! We’d be stuck right off the bat: there’d be no way to complete any of those tasks since they’d all indirectly depend on each other. A graph that is both directed and acyclic (and therefore free of these problems) is sometimes called a DAG for short."