Assignment 3: Why Vis

by Mike Gleicher on January 27, 2017

For week 3, we are hopefully in the “regular groove” for class. A reading due at the beginning of the week. A discussion that you are supposed to begin at the start of the week and continue through the week (and possibly into the next). An a seek and find.

Reading: The reading (try to have it done before class on Monday 1/30) is here.

Seek and Find: As usual, due on Friday (2/3) (link)

Discussion: this week’s group discussion on Canvas. If you aren’t assigned to a group, contact Chih-Ching. Please make an initial posting by Monday 1/30 (that means by the end of the day), and continue the discussion throughout the week.

In the past, I haven’t given specific questions for this reading. Students seem to find interesting things to discuss for this without much prompting.

But, if you’re stumped for ideas: In your initial posting, give your thoughts on the various perspectives we’ve read. You don’t need to summarize the readings- but you can mention what you found surprising, interesting, or what resonated with you (or didn’t).

Some comments on discussions and readings…

by Mike Gleicher on January 27, 2017

While I am not always contributing to the discussions, I am reading a lot of them. There are some great conversations happening. Hopefully, this is thought-provoking (and fun) for you. These first few discussions tend to be a little more philosophical than once we get to more technical material later.

(I think Chih-Ching will also be reading and contributing more in the future – after the qualifying exam Monday).

But something that I see over and over (and I remember from last time as well):

A lot of the readings are book chapters. They aren’t stand alone articles. I forget how much of a difference this makes. For example, the Munzner reading for this week is the first chapter of a textbook – it is meant to read like a textbook. Cairo and Tufte are book chapters as well. I think you miss out when you see a chapter from a book without seeing the whole book.

We’ll be tweaking the groups next week, now that the class enrollment has settled. Then we’ll probably try to keep the same groups for a few weeks (there’s a tradeoff – you get know people, which helps for discussions, but it’s good to mix things up every once in a while).

Also, I note that not everyone is contributing to discussions. The discussions are an important part of the class. If you’re having technical problems that are preventing you from contributing, talk to Chih-Ching or me.

Some pointers to postings…

by Mike Gleicher on January 25, 2017

The course schedule is being populated slowly. But it’s a place to see assignments before they happen.

Reading 3 has been posted but it isn’t easy to find yet. Seek and find 3 and assignment 3 are delayed a little – we want the class roster to stabilize before doing Canvas set up, and want to set up Canvas before making the content postings.

There is a lot of relevant stuff on last year’s course web. For example, my “4 steps to do vis” is “written up” as a posting “How to do Visualization.” There’s also some advice about learning D3 (and other toolkits) called “but I really want to learn D3“.

Books

by Chih-Ching Chang on January 25, 2017

If you are interested in reading more data visualization books/chapters in addition to the reading assignments, we provide some books/chapters in the protected reader under “Books” folder for you now. Check them out!

Reading 3: Why Visualization

by Mike Gleicher on January 22, 2017

Due Date: preferably before class on Monday, January 30th. You’ll get more from the lecture if you’ve done the reading. The associated discussion has parts due that day. (link to discussion assignment)

In this reading, we’ll explore the “why visualization” question a little bit deeper. Again, we’ll get 3 very different perspectives:

  1. Chapter 9 of Visual Thinking (course reader [Ware_visual_thinking_ch9.pdf; 18 pages] / library) by Colin Ware. Yes, we’re reading the last chapter first. You might want to skim through the book leading up to it (I basically read quickly) it in one sitting. Reading the ending might motivate you to read the whole thing (which we will later). The perspective here is how the perceptual science might suggest why vis is interesting. Remember that you can access this book online if you don’t have it.
  2. Chapter 2 “Visual Statistical Thinking” from Tufte’s Visual Explanations (pages 26-53; 27 pages) . The perspective here is historical – what can happen when Visualizations work or fail. A scan of the chapter is here. (3-VE-2-Visual-Statistical-Thinking.pdf; note this was a link to the wrong chapter originally)
  3. The first 17 pages of the Introduction to “Information Visualization: Using Visualization to Think” by Card, Mackinlay, and Schneiderman (01-InfoVis-CardMackinlaySchneid-Chap1.pdf). This is a 1999 book that consists of this intro, and a bunch of seminal papers. The examples are old, but the main points are timeless. It is the best thing I know of that gets at Vis from the cognitive science perspective. The rest of the chapter (past page 17) is good too, but more redundant with other things we’ll read – so it’s optional.

There’s a 4th reading – it’s more part of the next week (but there will be too much reading already next week), and I want to use it in class on Wednesday Feb 1. I want people to learn to critique things. Usually, we just critique – but one of my goals in this class is to teach people to do it more effectively. This chapter (which is part of a whole book on how to critique productively) will hopefully give you some things to think about, although ultimately, I think it just takes practice.

  1. “Understanding Critique,” Chapter 1 of Discussing Design by Adam Conor and Aaron Irizarry, O’Reilly Books, 2015. Chapter available online as a sampler from the publisher. (pp. 7-25, 18 pages)

If you missed class the first week…

by Mike Gleicher on January 21, 2017

Try not to miss any more classes 😉 (read the policy)

But, realistically we understand people miss classes (in fact, I missed the first classes). Especially people admitted off the waiting list.

So, if you missed the first week:

  1. If you’re going to drop (of course, I hope that you don’t drop), please drop sooner rather than later – there are people on the waiting list and we can’t let them in until spaces open up!
  2. Please do the online assignments – you are still responsible for the first week’s assignments. We are a little more lenient with deadlines in the first week or two. (reading 1, assignment 1, seek and find 1)
  3. Please try the “in-class exercise” – you can’t turn anything in, but you can learn from it.

In Class Assignment 1: The Simplest Visualization Problem

by Mike Gleicher on January 21, 2017

Note: this is meant as an in-class assignment. Students should not read the assignment before class. Part of the idea is to come at the problem fresh.

If you missed class, try the assignment on your own. It’s a lot less fun without a group. However, (after you try to come up with your own designs) you can use the original design exercise (here) to get some additional designs to look at.

The idea of this assignment is to explore the diversity of visual designs, to see how different visual designs might “group” together, and to start to think about what designs might be good for (or not). In order to do this, we will use a really simple “data problem.”

This exercise is inspired by one in a design class, described here. But don’t look at that page! It has 45 different answers. I want you to try to invent your own. (you can look at the page AFTER class). Also, this exercise was intended to focus on creating designs: not necessarily organizing or evaluating them.

We used a variant of this exercise in a workshop that I ran (with designer Stephan Thiel) to teach literature scholars about visualization. They turned out to be quite creative.

 

The Challenge and Assignment Overview

Your problem: design a visualization to communicate the numbers 75 and 37.

I’ve already given you one visualization (write out the numbers). How many more designs can you come up with?

The assignment will be divided into 3 phases. We’ll do all in class…

Phase 1: Generate Designs

This will be done in pairs. Each pair of people (if you can’t pair up, you might need to move to a different table – there should be at most 1 group of 3) needs to start with a blank piece of paper. Write both names on top of the page. Divide the paper into a 4×2 grid (we’ll give you suitable paper).

The goal of this phase is to generate as many different designs as you can. Make sketches. If you can’t sketch accurately enough, add text to explain what you are trying to draw.

The key here is to generate as many different designs as you can. The posting that inspired this exercise was called “45 ways to compare 2 quantities” (so you can guess how many that designer came up with). In the workshop, we came up with almost as many (and they were different than what were in the posting).

Don’t judge how good/bad they are (yet). Just make the designs.

To start things off, we’ll show you a few. Don’t copy ours.

Each pair of people should come up with at least 10 designs in 15 minutes (the clock is ticking). You can fit 8 on each side of your paper (see why we divided the paper).

Part of the goal here is to introduce the idea of design ideation: come up with many ideas, and then assess later. Part of the idea is to get you to realize there are lots of choices we can make in how to present data.

Each pair will turn in a page of designs (so make sure your name is on the page!)

Phase 2: Gather and Group Designs

With 30+ groups each generating 10+ designs, we’ll have a lot to look at. In this phase, we’ll try to group them together.

We’ll start as a class. We’ll pick one design. Then we’ll ask to see who else has this design, and who has variants. Then we’ll ask someone (pick a table) to give a design that is as unique / unlike the ones we’ve seen as possible. We’ll do this a little as a class, then we’ll let each table to this.

At the end, each table should have a set of “buckets” of common designs.

Part of the goal here is to get a bigger appreciation of the space of designs. Another goal is to see how to group similar things: what are fundamentally different designs, and what are just minor variants?

Phase 3: Evaluate Designs

In this phase, we’ll look at designs and ask “what could this be good for.”

For each design, try to come up with at least 1 thing that it is good for – what is a question you can answer quickly looking at the visualization. Then come up with (at least) 1 thing that it is less good at. For that 1 question that it isn’t good at, find some design that is better at that task.

We’ll do a few of these as a class, and then let people try doing it themselves in small groups. We’ll trade papers (each table trades with another table). Divide each table in half.

Part of the goal here is to start to appreciate how design connects to task, and to see the range of tasks. Part of the goal is to start to think about critique, especially in response to task.

Each half table should turn in a sheet with: everyone’s name on it. A sketch of each design you evaluated along with a list of tasks its good for, and a task that it isn’t good for.

Why are we doing this?

Hopefully, this exercise will get your creativity going in thinking about visualization, and help you start to realize the wide range of possible ways to do things. It should motivate the need for some terminology for discussing visualization, and for some systematic ways to define and describe designs. It should get you thinking about tasks, the connection between design and task, and critiquing with respect to tasks.

This will exercise will also hopefully be a fun ice-breaker. And get you used to the idea of creating and collaborating in class.

In terms of evaluation: we won’t be judging your artistic skill or design talent. Hopefully, you will be inspired to try as best you can (the more you try, the more you will get out of the experience). If learning doesn’t inspire you, maybe peer pressure will. Pretty much, everyone will get a “good” for this assignment (unless you obviously don’t try in all the phases – but we expect everyone to get a “good”). We will have no way to give anything better than “good” – but remember, getting a good grade in this class is about consistently participating in the exercises.

The Week in Vis: Week 2 Jan 23-27

by Mike Gleicher on January 21, 2017

This is the first (of 14) “Week in Vis” posts. The idea is to give you a sense of what is happening next week (and a brief reminder of what we’ve done already). The idea is to have a single posting that points you towards the important things you need to know about (since they can be scattered about on the web page). I guess, these posts are necessary since I’m not a good enough web designer, so I need to give guidance. You can also look at the blue box at the top right of the course web page.

Normally, these posts will be made on Thursday or Friday, to give you enough warning for the following week that you can be prepared. This first one is late, I apologize – and will push back any deadlines accordingly.

Last week (Jan 18-20), we had some stuff to get started. Everyone should have done the first seek and find, and the first discussion assignment. If you haven’t done these things yet, please do them soon.

And remember… you can still contribute to the first week’s discussions (the assignment and the seek and find). There were lots of interesting things that were coming up – so hopefully you all have things to talk about with each other!

We are still dealing with enrollment and waiting list issues. Hopefully this will stabilize this week. Because there are delays between waiting list processing, enrollment, and getting things enabled in Canvas, there can be confusions. Please bear with us – hopefully all the kinks will be worked out quickly. If something isn’t working as it should (e.g., you can’t access Canvas or aren’t in a discussion group) please let Chih-Ching know.

Next Week (Jan 23-27)… We’ll still be dealing with enrollment and waiting list issues. Hopefully these will all be resolved quickly. But as people drop the class, we can add people from the wait list, and see if they accept, and … Also, we’ll hopefully have more practice with Canvas, so that it will work more smoothly.

  • Reading 2 – usually readings are due before Monday, but because the assignment is going out late and because it’s a big reading, please read everything before class Wednesday Jan 25th.
  • Assignment 2 – includes a discussion related to the reading. Again, an initial posting is due on January 25th, and we expect people to continue to discuss the topic for a week.
  • Seek and Find 2 – it seems most people got the idea of the seek and find. They get more fun as we get more practice.
  • Class periods – we’ll start with our “regular” Monday/Wednesday schedule.
    • Monday, I’ll try to see what happens when I give a “traditional” lecture. Not sure how it will work in the weird room. The idea is a summary of key points of the class in 1 lecture).
    • Wednesday, there will be another “lecture” – informed by how things go on Monday.
    • Friday will be an “optional non-lecture” – my plan is to come to the class room, and just talk informally to whoever is there about visualization and the class. If no one comes, it will be a short conversation (or Alper, Chih-Ching and I will find things to talk about with each other). Ideally, it will be a chance for me to get to know some of you, and to hear about what you’re interested in.

I am hoping that by the following week (Jan 30-Feb 3), we’ll have things in the regular groove, with readings and assignments posted ahead of time, no more enrollment or Canvas issues, ….

Assignment 2: What is Vis?

by Mike Gleicher on January 20, 2017

This is a more “usual” weekly assignment – it has a reading, and a discussion. Also remember, that each week, there is a seek and find assignment. (for this week, there is Seek and Find 2).

Please make sure that you’ve done all parts of Assignment 1. (set up Canvas, the 1st reading, the 1st discussion). You can continue the 1st discussion even while Assignment 2 is in progress.

Reading: Reading 2 (best done before class on Monday, January 23rd, but since the assignment is going out late, Wednesday January 25th is OK)

Discussion: (initial posting due 1/25, please make a few responses to others to start the conversation by 1/27, and have some conversation by the time the discussion closes on Friday 2/3).

Initial Posting: What is your perspective on what visualization is, and how do the various perspectives of the different authors relate to your view?

Discussion: you’ll probably have a range of views (with the students and the reading authors) to compare, contrast and discuss. If nothing else, you can always discuss the differences in viewpoints between different authors – but hopefully the personal perspectives will come into play as well.

Everyone needs to make a minimum of 2 postings beyond their initial posting. It’s more fun and interesting for everyone if it is a discussion so I hope everyone contributes early and often.

As mentioned in the instructions for the first assignment, the discussion on Canvas will have the class broken into groups of approximately 8 people. We’re still working the kinks out of the canvas groupings – if you have a problem, please contact Chih-Ching.

Seek and Find 2: Why Vis This?

by Mike Gleicher on January 20, 2017

This second seek and find assignment has a little more to it than the first. Because we’re setting the assignment out late, the due date is later.

Due: Friday, January 27th. The assignment will remain open for late submissions until Friday February 3rd.

Turning it in: There will be a discussion on Canvas

Assessment: We will give you a score on the 70 point scale (see grading). A good assignment will include a valid picture and link, and a brief answer to the question.

What you need to do:

If you need the general directions and rules for a seek and find, see Seek and Find 1.

For this seek and find, you need to find a visualization for which you can say why this visualization was made. What is the task that required a visualization? Why couldn’t this have been just done with a few numbers or a little text? Why did the designer bother to make a visual representation?

Include a picture of the visualization, a link to it in context (if there is one), a brief description of what the visualization is, and an explanation of why it needed to be a visualization (those 3 questions in the previous paragraph are a good start).