The fundamental question that you have to be able to answer over and over again as a visualization designer is how do I visualize this? How do I go from data to a visual representation that helps you answer some important questions that you have about the data? So typically, what happens is that you receive some dataset and this dataset, let's say that it is in form of a table, typically is in the table format, contains a lot of information, a lot of attributes. And one of the first things that you have to figure out is how do I transform all this information into a visual representation? This is a very complex step. So, one way I like to describe this is that we can break the problem down into two steps. The first step is to decide what to visualize. And the second step is about deciding how to visualize it. These two things are of course highly related, but it's very important to keep in mind that these are two separate steps. So, one way I like to describe these steps more specifically is that the first one is about selecting and transforming the data. What do I mean when I say that? I mean that originally, typically, you receive a table with a lot of attributes. But every single representation that you create is typically created by using a very small subset of the attributes that you have in the original table. Another thing that happens basically all the time, is the data needs to be transformed in order to be in a format or in order to extract the information that you need in order to create the chart that solves the problem that you want to solve. So, I like to call this select and transform. The second step is about choosing or designing an appropriate representation. Why did I put choose and design? Because choosing is exclusively about selecting an existing format that is appropriate for the goal that you have in mind. What is designing is more in general trying to create a novel representation for the problem that you have. So now, in the following videos, I'm going to address step two first, and then we are going back to step one. Why do we do that? Because it's easier to first show you a number of predefined fundamental graphs and what kind of data they can accommodate, and what kind of information they can communicate. And then show you how some transformations of the original data is needed before the data can be visualized with these graphs.