Now, in this last video, I want to recapitulate a little bit what we covered in this module. So, we started with talking about interaction, what is the value of interaction and what kind of benefits it can bring in a visualization and in a visualization tool. The main idea is that, you by using interaction, what you can do is to help a person answer multiple questions using one single visual representation. That's a very, very powerful aspect of visualization. In particular, you can do that by manipulating different aspects of a visualization. We talked about the information visualization pipeline and how these aspects pertain to different transformation steps in this pipeline. These three aspects are data, so you can change or manipulate the data through interaction; you can change or manipulate the mapping; and you can change the actual final view, the visualization itself. So, this is very useful to keep in mind. Then, we went through a very long list of interaction methods, which we organized around these three main classes of changes, so changing the view, changing the mapping, and changing the data. So, within changing the view, we covered selection, navigation, and spatial arrangement. These are three things that you can change using interaction that affect the main view. The second one is change mapping. And with change mapping, we intend changing the actual rules that transform data into visual properties. Finally, we have aggregation and filtering, which can be manipulated interactively and affect the data, and as a consequence, also affect the final visualization. After covering this part, we started talking about the choice between single view and multiple views. We discussed what are the benefits of using multiple views. Well, the main idea behind multiple views is that, by using multiple views, you can represent different aspects of the same dataset at the same time. In particular, what you can do with multiple views is, you can present in these views different information and or information in a different way. So, these are the two main classes of methods that you can use to represent different facets of a given dataset. So, either different information extracted from the same dataset and/or different visual representations for the same information. I also covered, finally three main patterns that you can be faced with. These scenarios are very common ones, and all of them can be tackled with multiple views and interaction. One last note that I want to say before concluding is that every time you are designing a visualization tool or application that uses interaction and/or multiple views, it's very important for you to ask yourself, what is it that you need? Why are you using interaction and multiple views. So, in other words, don't just use interaction and multiple views because you think it may be nice. I think what is really useful is to acquire this mentality, by which, every time you use any of the methods that we introduced in this module, you also ask yourself what is the benefit of using that. So, what is the benefit of adding this type of interaction and what is the benefit of having multiple views? Why am I saying that? Well, because I've been talking mostly about the benefits of using these methods, but there are also some costs, in particular interacting with the visualization as a cost because interaction is very slow compared to just observing something with your eyes. So, that's one problem. Splitting information in multiple views also has a cognitive cost. It means that you have to look at different elements on the screen, but you also have to mentally integrate information that is spread across multiple views. So, don't make the mistake of thinking that using Interaction and using multiple views is always beneficial. On the contrary, it's always a tradeoff. So, it's very important for you, let me tell- let me say this again, every time you are employing some interaction methods or some multiple view methods in your visualization design, ask yourself, what are the benefits and what are the possible negative consequences and try to convince yourself, if you actually want to use any of these methods, you have to convince yourself that this is ultimately going to be beneficial.
