[MUSIC] Welcome back. Now, let's consider a fundamental question for any visualization. Is the primary intention to highlight something you've already found in the data or perhaps to enable the discovery of things, as yet unknown? Where does it fall in the spectrum between exploration and explanation? These factors should help guide you in your design strategy. If the purpose is exploratory, then the visualization will often need to provide users with more direct control and interactivity for the analytic process and workflow and those users also probably come with prior knowledge about the data they are seeing. For purely explanatory purposes, interactivity may not be needed, nor any prior knowledge expected. In this latter case, there may not be any workflow at all, depending on the context and users, providing either too much or too little detail can also be counterproductive. You should recognize that simplicity can often be relative. [MUSIC] Along with thinking about key characteristics of your audiences, you should also consider the context and purposes for which they'll be using the visualizations. For example, as we have touched on previously, will the visualization be used to explore data or to convincingly explain something already discovered to other people or perhaps some combination of both? These factors will help guide your design strategy. Please take a look at this graphic. That offers when general framework to think about the context of the visualization. If your primary audience's goals fall on the exploratory side of the spectrum, then the visualization and associate interface should generally offer a fair degree of user control that may include interactive elements like filtering, pivoting and zooming, which we'll discuss further in a later lesson. The trade off here is that you're asking more of your users both in terms of effort, knowledge and motivation to find answers. But in the case of the high user control group, you might expect them to bring more prior subject matter expertise to the table and motivation to look for answers. On the exploratory side of the spectrum, it may be something as simple as making a case or statement in an infographic to an audience that has no prior subject matter expertise. In that case, basically, it all comes down to the clarity of message. The more quickly and easily you can convey a point, the better. Open-ended discovery and iterative investigation are not the point here, it's more of a once and dunce thing. Does the presentation make any key takeaways clear, crisp and compelling? Another related factor included in this graphic is the relationship between dynamic versus static data visualizations. That is how much or little interactivity is there, or the ability to change views by users? A printed infographic, for example is clearly a static or a read only visualization. The viewer can do nothing, but see and consider what is being presented. On the other hand, a network visualization in say, a cybersecurity investigation tool can be highly interactive with all kinds of options for users to expand, hide filter. And otherwise, modify what the visualization is showing. In this latter case, the user's direct engagement with the visualization is necessary to reveal stories in the data. Dynamic visualization has a steeper learning curve and static, but that may be exactly whats needed for the context. Of course, you can create static visualizations that have a lot of complexity and depth for people to study and explore. But in general, these days, more complex and deep visualizations benefit from interactivity at least on some level. Just because you can show a lot of data all at once, it doesn't mean you should. On the other hand, you don't want to throw the baby out with the bath water when decluttering a visualization. One way to do that is to consider the use of progressive disclosure. That is showing things only as needed based on factors, such as context, urgency and criticality. In other words, only show the amount of data that's useful at a given stage of a use-case or scenario. For example, let's say, you're looking at the status monitor for a nuclear power plant. You could display everything that's being monitored at the same level of visual hierarchy. That is every system that's functioning properly would be displayed exactly the same way, as the systems that aren't. A potential problem with this approach is that any malfunctioning system indicators may get lost amidst all the dials and controls, and visualizations for the properly functioning systems. An alternative approach for a system monitoring display that helps alert users to problems as quickly and easily as possible is to design a view that shows data about something that has surpassed a level of criticality with a larger, and more detailed visualization while condensing, and summarizing the things that are running normally, and properly. Of course, it can be possible to access the data about the normally running systems with a little additional effort, but only the immediate need is shown in a more obvious and immediate way as is represented in this simple example. Now, recall that simplicity is often not a simple matter. The level of information presented and amount of control, given to users depends on who they are and what they need to do in a particular situation. As the head of Apple design, Jonathon Ive puts it, simplicity is not the absence of clutter. That's a consequence of simplicity. Simplicity is somehow essentially describing the purpose and place of an object, and product. The absence of clutter is just a clutter free product, that's not simple. Now in this lesson, we've seen that there's a wide spectrum of potential purposes for visualizations and the nature of those purposes shapes the direction of what you design. Are you helping people to discover unknowns in the data or to present a compelling case for something already discovered? Have you designed the visualization with the appropriate levels of detail and description to fulfill the purpose? Basically, is the design simple, but not simplistic for the need? All of those things need to go into your considerations, as a data visualization designer. See you next time.
