The Week in Vis: Week 03 (Sep 20-25)

Our second full week of class. Hopefully, you’ve gotten used to to the pattern. This week, we’ll introduce the final piece of class: design exercises and challenges. These will be due on Mondays (so I am warning you about something for the 27th).

The topic for this week is abstraction: how do we talk about data and tasks in general ways. This will allow us to get away from the details of any specific problem, and to transfer our knowledge between problems. It will also allow us to consider general purpose tools. We’ll also talk about critique (which you read about last week).

This week’s driving question is “How do we talk about Visualization / think about Visualization?” - which leads to the question of abstraction - how we consider visualization in general ways.

I phrase the question this way to emphasize a point: we want to use abstraction as a tool to talk about visualization. A big goal is to develop a vocabulary to discuss visualization in a way that is independent of its details. Especially with task abstraction it is easy to get nitpicky about how different people like to define things. Instead, my goal is to develop some shared vocabulary for talking about task and data.

There are two parts here: data abstraction and task abstraction.

Data abstraction is (relatively) straightforward. But there are useful terminology and distinctions that will prove helpful as we start to connect data to visuals.

Task abstraction is much murkier. There is no one right answer. There are many different ways to describe and categorize tasks - many of which will prove useful. From the readings and class activities, hopefully you will get to see many of them.

This week, the main topic will be introduced in the readings. There is a lot to read this week. I am aware of that - but it’s all good and important stuff. Because I want to do a discussion and exercise about critique, we’ll only take one lecture for discussing abstraction. Usually, the basics of data abstraction is easy for people, so we won’t need too much time in class. And, it will come up as you start to use Tableau…

Tableau is a segway to talk about another class element: Design Exercises. This is my term for things you have to do at home to “practice” visualization. In the past, they’ve been grouped together into a small number of “Design Challenges” - but this year, I am going to try to divide them up. If you prefer, you can think of this weeks exercise as “Phase 1 of Design Challenge 1” (look for it last year).

We are going to ask you to use Tableau (a commerical visualization tool) for some future design work. We will give you access to it (see the Tableau page). This week, your Design Exercise 1 is basically asking you to get started with it. Make some visualizations. We don’t care much about what they are.

Tableau is actually relevant to the topic of the week: it embodies these notions of abstraction to allow it to infer appropriate ways to present data. It is also a convenient way to produce visualizations from real data sets, which you will do in the next few weeks.

As you will notice, Design Exercises will be due on Mondays (so this one is due next Monday, the 27th). Generally, the cutoff for doing them will be the day before the next one is due.